High performance electronic message delivery engine
In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server and which recipient is interfaced to the Internet through a recipient server, where the message includes at least recipient information and message body, is disclosed. The method includes testing the recipient information by establishing a temporary test connection through the Internet from the sender server to the recipient server, before relying on the recipient information for sending the message, so as to provide an indication of deliverability of the message to the recipient without using an SMTP VRFY command.
This application claims the benefit of U.S. Provisional Patent Application No.60/485,167 filed Jul. 7, 2003, the contents of which are hereby incorporated by reference.
BACKGROUND OF THE INVENTIONThe present invention relates generally to electronic messaging systems and, more particularly, to an electronic message delivery system for processing electronic messages from a sender to a recipient through the Internet.
Electronic Message Delivery Challenges
Electronic messaging has quickly become an integral part of everyday life in modern society. Today, electronic mail (e-mail) is the most widely used application of the Internet and the fastest growing communication medium. In addition to the dramatic expansion of one-to-one user correspondence, many companies have come to realize that regular outbound electronic message communication, as well as automated customer care services via e-mail, can strengthen a company's relationship with its customers. However, sending and receiving high volumes of e-mail presents a number of technical challenges such as, for example: 1) management and support of ballooning infrastructures for high volume sending; 2) maintenance of security on an application that is open to the Internet; 3) the need to apply virus, content and other filters to e-mail as it enters or departs the network; 4) management and prioritization of message queues; 5) the use of native customer data to trigger, target and define messages to customers; 6) management of inbound “bounced” e-mail; 7) the ability to consistently and automatically support customer preferences; and 8) executing delivery quickly and efficiently.
Meeting these challenges for scalable electronic messaging programs has been difficult, if not impossible, using available technology and has required substantial investment in hardware and human capital. Currently, companies and organizations of all types, including e-mail specialists and service providers, continue to deploy decades-old technology for sending and routing e-mail across the Internet via Mail Transfer Agents (MTAs) that did not anticipate the explosion in traffic volumes and system complexity of today's networks.
Prior-Art MTAs and Open-Source Mail Routing Servers
E-mail was developed to deliver electronic correspondence from one user to another on a computer network, where a user is a member of an e-mail messaging system and may be a person, a machine or an application having an e-mail address within the domain of the e-mail messaging system. The primary technology for e-mail is the TCP/IP communication protocol known as Simple Mail Transfer Protocol (SMTP), which was invented in 1981 and is designed to support the routing and delivery of electronic messages from a user of one computer to a user of a receiving computer via compatible routing delivery servers across the Internet. SMTP facilitated e-mail delivery by standardizing message formats with common headers and routing information and establishing command syntax and processes for communication transactions between e-mail handling applications.
In general, e-mail applications fall into at least one of three classifications that play a specific role in the process of moving and managing e-mail messages. The first application classification is a Mail Transfer Agent (MTA), which transfers e-mail messages between hosts and networks across the Internet using SMTP. A message may travel across several MTAs on its way to its intended destination. The second application classification is a Mail Delivery Agent (MDA), which transmits outbound messages to the MTA for delivery from the local host or network, and is invoked by the MTA to file incoming e-mail in the intended recipient's mailbox. In practice, many MTAs have an MDA built in, and some MDAs include certain MTA components. The third application classification is a Mail User Agent (MUA), which is essentially an e-mail client application that, at minimum, allows a user to read and compose e-mail messages from an MDA.1
In the current state of the art, MTAs route e-mail messages from local hosts and networks through the gateway to the Internet to other MTA mail servers en route to recipients. MTA technology was developed by Eric Allman at UC Berkeley in the mid 1980's, and the basic architecture and components of the original MTA technology are still being used today.2
MTAs—Current Art
In the current state of the art, the standard process for MTA e-mail sending is as follows:
-
- 1. At the MTA, accept an incoming SMTP-formatted message from an MDA or other e-mail generating application;
- 2. Write the incoming message to a file on the hard drive for temporary storage;
- 3. Read the message's SMTP headers for destination information;
- 4. Find the destination's mail exchange (MX) record through a Domain Name Service (DNS) lookup;
- 5. Open a connection (i.e., socket) with an MTA at the destination network;
- 6. Through this socket, conduct an SMTP communication consisting of a series of information transmissions from the sending MTA and replies from the receiving MTA, verifying identities and validating deliverability or communicating temporary or permanent deliverability failures;
- 7. Transmit the message; and
- 8. Await any short-term reply messages from the recipient MTA containing delivery failure errors (“bounces”). In the event of a delivery failure error, resend the message. Otherwise, delete the message from the hard drive and end the process.
The sequence from connection to transmission is substantially uniform and standard in the current state of the art. The transmission of the message is the first part of a complete delivery process in which, after the message is transmitted, the MTA continues to store a copy of the message for a period in which any short-term failure messages (i.e., bounces), SMTP e-mails sent to the sending MTA from the receiving MTA communicating a delivery failure, may be received. If such bounces are received, the sending MTA uses the stored copy of the message to repeat the entire process and send the message again (i.e., retry). Typically, an MTA will retry a bounced message several times in succession before terminating the delivery process and forwarding a delivery failure message to the sender. This process may repeat while an e-mail is passed between MTAs en route from its origin to its final destination.
Turning now to the figures, in which like items are indicated using like reference numbers, attention is initially directed to
Referring now to
Continuing to refer to
Referring now to
To summarize, the key processes in e-mail MTAs in the current state of the art are as follows:
-
- 1. SMTP feed: the exclusive receipt and processing of pre-formed SMTP messages;
- 2. Store-and-forward: the mechanism of writing each message to a file on the hard drive for temporary storage, forwarding the file to the receiving MTA for delivery and subsequently deleting it after the completion of the delivery process to ensure recovery, and redelivery, in the event of a system failure before the delivery process had been completed;
- 3. Combined validation and transmission: a fixed sequence of communication transactions within the SMTP communication with recipient MTAs that culminates in message transmission;
- 4. Relay: a delivery process that sends each message it receives and continues to resend those that fail; and
- 5. Closed architecture: an architecture that combines the aforementioned elements in an uninterruptible sequence of processes.
Recent Advances in the MTA Art
The MTA processes and characteristics described above have not changed fundamentally since the first modern MTA application, sendmail, was released in 1981. To date, advances in the art of MTA e-mail delivery have not altered any of the fundamental processes and characteristics as described above, but have, rather, worked within their constraints to optimize delivery performance by improving the speed and efficiency by which these processes are performed.
To such end, recent advances in the art have developed in conjunction with advances in the functionality of hardware operating systems that support certain of the existing key processes for MTA e-mail delivery, namely, the generation of socket connections to recipient MTAs and the storage and retrieval of message files to and from the hard drive, or file input/output (i.e., file I/O).
The generation of socket connections is the foundation of SMTP mail delivery. In the earliest embodiment of the art, sendmail, the method for establishing a socket connection to a recipient MTA relied upon the creation of a new instance of the application for each new socket (i.e., process-based socket connection). This approach is inherently processor intensive and limits delivery speed to the rate at which the application can instantiate itself on the hardware.
Thread-Based SMTP Connections
As e-mail volumes have increased, and the practice of sending messages to large lists of recipients for publishing, marketing and other purposes emerged, interest in improving the speed and processing performance of e-mail MTAs has intensified. At the same time, advances in operating systems provided the ability to support multiple concurrent chains of commands within a program (i.e., threads), enabling concurrent processes in support of multiple simultaneous data transactions with substantially lower processing overhead than required for instantiation. Thus, an e-mail MTA could generate multiple socket connections simultaneously, enabling the parallel processing of SMTP connections for much faster delivery with much lower processing overhead (i.e., thread-based socket connection).
The open-source MTA application qmail is the earliest and most widely used example of this method. Subsequent improvements in threading models for various operating systems provided more efficient memory allocation for threading that enabled dramatic increases in the number of threads that could be efficiently used for socket connections. This increase in the number of threads led to the emergence of qmail-type MTA applications optimized to run most efficiently on operating systems configured for massively multi-threaded sending of electronic messages. The most successful examples of this advance are the commercial products PowerMTA™, released in 1999, and Lyris MailEngine, released in 2001.
Asynchronous File I/O
Improvements in speed and performance arising from multi-threaded SMTP communications highlighted a new rate-limiting process in MTA e-mail delivery, the process of writing and retrieving message files to and from the hard drive, file I/O. File I/O processes are common across computer applications, and the propensity of the speed of an application, or series of processes, to slow down or bottleneck behind the sequential reading and writing of files to disk is widely known. A first significant advance in this process introduced the management of data storage across multiple drive discs in parallel (RAID), dramatically increasing speed. A second, more recent, advance in this process, released in open-source operating systems OpenBSD and Linux RedHat, enables the operating system to read and write multiple files to and from a drive simultaneously in increments, or asynchronously (i.e., asynchronous I/O). This advancement enables much more rapid reading and writing of files to disc, creating a marked improvement in efficiency for the file I/O processes in MTA applications, thereby improving processing speed and efficiency.
Thus, it became possible to optimize a qmail-based MTA application on an operating system that supports massive multi-threading and asynchronous file I/O to achieve the best performance to date from an e-mail MTA application. In fact, to ensure the optimum configuration of the operating system and tuning of the MTA application, the preferred embodiment of such an application is a pre-configured hardware appliance with RAID. Two such appliances were launched in 2001-2002: IronPort A60™ and Mirapoint Message Server. A comparison of the sustained performance of the prior art MTA applications described above, according to the published specifications, is summarized in TABLE 1.
The most common performance metric for MTAs is the sustained rate at which they transmit standard text messages (ranging from 5,000 to 10,000 bytes) from a standard dual Pentium 3 or Pentium 4 processor server, or an equivalent thereof. The Sendmail MTA performs the most poorly, as its speed is impaired by instantiation for socket connections and standard file I/O processing. Standard implementations of qmail perform about three times better, through the use of thread-based socket connections. Qmail-based multi-threaded MTAs offer a dramatic ten-fold improvement in performance over standard implementations of qmail while qmail-based multi-threaded MTAs that also use asynchronous I/O offer an additional 50% or more increase in performance over massively multi-threaded qmail-based MTAs without it.
Application Program Interfaces for the MTA
While sendmail and other MTAs evolved to use features and rule sets for content analysis, options available to users in performing modifications to SMTP messages and other complex processing (such as, for example, spam and attachment filtering, custom logging, disclaimer addition and virus scanning) have been limited to: 1) calls to external programs, which are generally usable only for local delivery or messages to local addresses; 2) use of the MTA in queuing mode with a program that is triggered on a regular basis; 3) use of a forwarding scheme (i.e., serial architecture with several MTA instances); or 4) altering sendmail's source code to add some extensions, thereby requiring recompilation of sendmail.3 None of these four options are simple to use or efficient in operation. Lacking common interfaces to other application, this architecture is also referred to as a “closed architecture.”
In order to facilitate the writing and integration of processing filters for e-mail, the sendmail team developed some application program interfaces (APIs) for enabling the introduction of external e-mail processing sequences at various steps in the SMTP process of the MTA.4 These APIs were collectively called MILTER (Mail Filtering API). MILTER was first implemented in version 8.10.× of sendmail and has been officially supported in sendmail since version 8.12.×.
The MILTER API is limited in scope to post-processing, in which an API event is triggered only after each SMTP event occurs during the protocol (e.g., Connect, HELO, MAIL FROM, RCPT TO, DATA and QUIT). That is, messages may be acted upon while traveling INBOUND into a server, when a receiving server is responding to the actions of a sending server. However, there is no provision for acting upon messages before an event occurs and, consequently, there is no facility for acting upon messages traveling OUTBOUND from a server when a sending server is initiating actions.
As will be seen hereinafter, the present invention provides a remarkable improvement over the prior art as discussed above by virtue of its ability to provide increased performance while resolving the aforedescribed problems present in the current state of the art electronic message delivery systems.
References
- 1. Bill Ball, Hoyt Duff, Red Hat® Linux 9, Sams May 8, 2003, Chapter 11.
- 2. David Pitts and Bill Ball, Red Hat® Linux 6 (Unleashed), Sams Jul. 30, 1999, Chapter 7 SMTP and Protocols.
- 3. Stéphane Lentz, “Milter,” http://milter.free.fr/intro/, slides 1 and 2, France, Apr. 14, 2004.
- 4. Sendmail, Inc., “Milter API,” http://www.milter.org/milter_api/api.html, 2003.
As will be described in more detail hereinafter, there is disclosed herein a method for use in a messaging system interfaced through the Internet. This method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, where the message includes at least recipient information and message body, includes testing the recipient information by establishing a temporary test connection through the Internet from the sender server to the recipient server, before relying on the recipient information for sending the message, so as to provide an indication of deliverability of the message to the recipient without using an SMTP verify (VRFY) command.
In another aspect of the invention, there is disclosed a method for use in a messaging system interfaced through the Internet. This method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, where the message includes at least recipient information and message body, includes confirming deliverability of the message to the recipient according to the recipient information. The confirming includes establishing a connection through the Internet between the sender server and the recipient server, and, from the sender server, using the connection to verify the recipient information with the recipient server so as to provide an indication, without using an SMTP VRFY command, as to whether the recipient is interfaced to the Internet through the recipient server. The indication is one of a positive confirmation that the recipient is indeed interfaced to the Internet through the recipient server and a renouncement showing that the recipient is not interfaced to the Internet through the recipient server.
In yet another aspect of the invention, there is disclosed a method for use in a messaging system interfaced through the Internet. This method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, includes, at the sender server, receiving from the sender a plurality of discrete data packets for use in defining the message. At least one of the plurality of discrete data packets includes recipient information and at least another one of the plurality of discrete data packets includes message body. The method further includes testing the recipient information by establishing a temporary test connection through the Internet from the sender server to the recipient server, before relying on the recipient information for sending the message, so as to provide an indication of deliverability to the recipient without using an SMTP VRFY command. The method also includes, based on the indication, performing one of (i) generating the message by combining at least the recipient information and message body then establishing a delivery connection through the Internet from the sender server to the recipient server followed by directing the message through the delivery connection to the recipient server for delivery to the recipient, and (ii) aborting the delivery of the message.
In a further aspect of the invention, a method for processing a message from a sender and intended for delivery to at least one recipient in a messaging system interfaced through the Internet is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a recipient server. The method includes, at the sender server, receiving from the sender a plurality of discrete data packets for use in defining the message. At least one of the plurality of discrete data packets includes recipient information and at least another one of the plurality of discrete data packets includes message body. The method also includes confirming deliverability of the message to the recipient according to the recipient information by establishing a connection through the Internet between the sender server and the recipient server, and, from the sender server, verifying the recipient information with the recipient server through the connection so as to provide an indication, without using an SMTP VRFY command, as to whether the recipient is interfaced to the Internet through the recipient server. The indication is one of a positive confirmation that the recipient is indeed interfaced to the Internet through the recipient server and a renouncement showing that the recipient is not interfaced to the Internet through the recipient server. The method further includes, based on the indication, performing one of (i) generating the message by combining at least the recipient information and message body, responsive to the positive confirmation, then directing the message so generated to the recipient server through the connection for delivery to the recipient, and (ii) aborting the delivery of the message, responsive to the renouncement.
In a still further aspect of the invention, a modified message server configured to replace an unmodified message server interfaced to the Internet for processing a message from a sender and intended for delivery to at least one recipient is disclosed. The sender is interfaced to the Internet through the unmodified message server capable of communicating with a recipient server. The unmodified message server includes a hard drive configured for saving the message as a message file thereon, and the unmodified message server is configured to forward the message file to the recipient server. The modified message server includes an arrangement for receiving from the sender as a plurality discrete data packets for use in defining the message. At least one of the plurality of discrete data packets includes recipient information and at least another one of the plurality of discrete data packets includes message body. The modified receiving arrangement includes electronic memory for saving the plurality of discrete data packets therein. The modified message server further includes an arrangement for processing the plurality of discrete data packets and generating the message by combining at least the recipient information and message body.
In a yet further aspect of the invention, a method for processing a group of messages from a sender in a messaging system interfaced through the Internet is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a plurality of recipient servers, each one of the group of messages being intended for delivery to a recipient. The method includes, at the sender server, receiving from the sender a plurality of discrete data packets for use in defining the group of messages, at least one of the plurality of discrete data packets including recipient information and at least another one of the plurality of discrete data packets including message body for a corresponding one of the group of messages. The method further includes generating the group of messages by combining at least the recipient information and message body for each one of the group of messages. The method also includes arranging the group of messages in an order in accordance with a predetermined set of rules based on at least one of the recipient information and the message body of each one of the group of messages.
In another aspect of the invention, a method for processing a message from a sender and intended for delivery to at least one recipient in a messaging system interfaced through the Internet is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a recipient server. The method includes, at the sender server, receiving from the sender a plurality of discrete data packets for use in defining the message. At least one of the plurality of discrete data packets includes recipient information and at least another one of the plurality of discrete data packets includes message body. The method also includes directing the plurality of discrete data packets through an application program interface to one of a plurality of external applications for performing at least one of (i) modifying the plurality of discrete data packets according to a first predetermined set of rules, (ii) adding additional data packets to the plurality of discrete data packets according to a second predetermined set of rules, and (iii) ordering the plurality of discrete data packets according to a third predetermined set of rules, and, consequently, for providing a plurality of resultant data packets. The method further includes receiving the plurality of resultant data packets from the one of a plurality of external applications through the application program interface, and generating the message by combining at least a portion of the plurality of resultant data packets. The method additionally includes establishing a connection through the Internet from the sender server to the recipient server, and directing the message through the connection to the recipient server for delivery to the recipient.
In a further aspect of the invention, a messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, and the message includes at least recipient information and message body. The messaging system includes a testing arrangement for testing the recipient information by establishing a temporary test connection through the Internet from the sender server to the recipient server, before relying on the recipient information for sending the message, so as to provide an indication of deliverability of the message to the recipient without using an SMTP VRFY command.
In still another aspect of the invention, a messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, and the message includes at least recipient information and message body. The messaging system includes a testing arrangement for testing the recipient information by establishing a connection through the Internet from the sender server to the recipient server, before relying on the recipient information for sending the message, in a way which provides an indication of deliverability to the recipient without using an SMTP VRFY command. The messaging system also includes a selective delivery arrangement for performing, based on the indication, one of (i) directing the message to the recipient server through the connection for to the recipient, and (ii) aborting the delivery of the message.
In a further aspect of the invention, a method for processing a message from a sender and intended for delivery to at least one recipient in a messaging system interfaced through the Internet is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a recipient server. The messaging system includes an electronic memory for processing data. The method includes handling the message as data in the electronic memory. As an additional aspect, the messaging system further includes a hard drive for storing files, and the message is handled in the messaging system without storing the message as one of the files in the hard drive.
In a still further aspect of the invention, a method for processing a message from a sender in a messaging system interfaced through the Internet is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a plurality of recipient servers. The message is intended for delivery to a recipient. The method includes, at least the sender server, separating the message into a plurality of discrete data packets, where at least one of the plurality of discrete data packets includes recipient information and at least another one of the plurality of discrete data packets includes message body. The method also includes testing the recipient information by establishing a connection through the Internet from the sender server to a recipient server, before relying on the recipient information for sending the message, in a way which provides an indication of deliverability to the recipient without using an SMTP VRFY command. The method further includes, based on the indication, performing one of (i) directing the message through the connection for delivery through the recipient server to the recipient, and (ii) aborting the delivery of the message. Alternatively, the testing is performed through a temporary test connection, while the message is directed through a delivery connection.
In a yet further aspect of the invention, a method for processing a group of messages from a sender in a messaging system interfaced through the Internet is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a plurality of recipient servers. Each message in the group of messages is intended for delivery to a recipient. The method includes, at the sender server, separating each message in the group of messages into a plurality of discrete data packets, where at least one of the plurality of discrete data packets includes recipient information and at least another one of the plurality of discrete data packets includes message body for the corresponding message out of the group of messages. The method also includes arranging the group of messages in an order in accordance with a predetermined set of rules based on at least one of the recipient information and the message body for each message out of the group of messages.
In another aspect of the invention, a method for processing a message from a sender and intended for delivery to at least one recipient in a messaging system interfaced through the Internet is disclosed. The sender is interfaced to the Internet through a sender server capable of communicating with a recipient server. The messaging system includes an electronic memory for processing data and a hard drive for storing files. The method includes selecting at least one of (i) handling the message as data in the electronic memory without storing the message as one of the files in the hard drive, and (ii) storing the message as one of the files in the hard drive.
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGThe present invention may be understood by reference to the following detailed description taken in conjunction with the drawings briefly described below. It is noted that, for purposes of illustrative clarity, certain elements in the drawings may not be drawn to scale.
The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a patent application and its requirements. Various modifications to the described embodiments will be readily apparent to those skilled in the art and the generic principles herein may be applied to other embodiments. Thus, the present invention is not intended to be limited to the embodiment shown but is to be accorded the widest scope consistent with the principles and features described herein.
Problems in the Current State of the Art
Advances in the art have yielded improvements in delivery speed and hardware efficiency in conjunction with advances in operating systems. There are, however, problems in a number of the key characteristics of e-mail MTAs that limit advances in performance, flexibility, security and feature enhancements.
Problems with the SMTP Feed
E-mail was initially created to transmit electronic messages from one (usually human) sender on a computer network to another. E-mail MTAs were created to receive and deliver e-mail messages to and from a local mail delivery agent hosting sender accounts to other MTAs outside the host or network. As the use of e-mail and the Internet have grown, users and organizations began to identify applications for e-mail outside of human-sender-generated correspondence, beginning with the use of e-mail for mass publishing and evolving into applications for direct marketing, automated transactions, support and customer care. These new applications offered tremendous potential for dramatically reducing the costs and improving the efficiency of mass-communication activities, and, thus accelerated interest in corresponding technology solutions.
The most promising applications for e-mail revolved around the ability to use databases of information on customers, members or subscribers to identify recipients, populate message content and trigger e-mail delivery. More fundamentally, an increase in the prevalence of interactions and transactions on web pages drove demand for e-mail receipts and other confirmations of online interactions.
Thus, it became important to enable databases and local web applications to send e-mail, but, since messages in this case would be created by system commands and not composed by individual users, standard MDAs would not support the required process, and e-mails would have to be sent from these systems to MTAs directly. MTAs, however, only accept pre-formed SMTP messages, while web and database applications generate data outputs. Consequently, system developers created SMTP “feeders” (e.g. Javamail) for local systems and purpose-built e-mail marketing and publishing systems that would convert data packets into electronic messages for transmission to local MTAs. Yet the SMTP feeder inherently limits the delivery speed of any MTA to the rate at which it can be fed by a feeder (in current art, about 200,000 e-mails per hour on a sustained basis).
Furthermore, as the uses of e-mail become more sophisticated, it is increasingly desirable to use information from two or more source systems to generate a message. For example, an e-commerce application may generate data for a text message containing specific information on a recently completed transaction. It may be desirable to draw targeted advertising or service content, based on information contained in the message, from a separate system. In the current state of the art, such combination of information from multiple source systems would require processing a transaction between the first system and the second system in advance of submitting a full set of content data to the SMTP feeder for generating the SMTP e-mail message. This processing between the first and second system implies significant integration between these systems, as has only been performed for large-scale e-mail operations, to the best knowledge of Applicants, by one company, Amazon.com.
Additionally, it may be desirable for an e-mail MTA to support other electronic message protocol formats in order to accept messages in other protocols for conversion to SMTP e-mail or to accept SMTP e-mail messages for conversion to other protocols, based on specifications within each message, centralized business logic or other mechanisms.
Problems with the Store-And-Forward Processing
Related to the practice of SMTP feed, is the current practice of store-and-forward processing in which the MTA accepts SMTP e-mail messages into the application, writes each one to a file on the hard drive, sends it on through the delivery process and then deletes the stored message from the hard drive. This practice creates a number of problems for e-mail delivery.
As discussed previously, writing files to the hard drive, file I/O, is long-standing problem in MTA performance. While recent innovations in asynchronous file I/O have improved performance, the activity of mechanically reading and writing files to and from the hard drive is inherently processing intensive and prone to bottlenecks.
An additional problem to store-and-forward processing is the sequential, first-in-first-out (FIFO) nature of message processing in the system, arising from the fact that each e-mail is processed in isolation of any others sent in conjunction with, or in proximity to, the one being processed. This leads to one of the most significant problems in high-volume MTA e-mail delivery performance. E-mails are queued in sequence, one behind the other, for transmission through active connections. In high-volume applications, messages are further grouped by domain. This presents the unfavorable circumstance in which SMTP communications are directed at the same domain simultaneously, increasing the likelihood of slowdowns or connection failures at recipient domain MTAs. When such connections are attempted and slow or fail, they become inactive. Multiple attempted connections to the same domain, created as individual messages to the same domain, travel through the application in sequence, inactivating additional connections and slowing or halting delivery across the application, despite the initial availability of very large numbers of threads.
FIFO processing associated with store-and-forward processing creates an additional limitation to e-mail sending. In the same manner as the system is unable to review multiple recipient addresses to manage delivery attempts as connections become inactive, it is also unable to review messages and groups of messages to prioritize traffic across a queue or set of queues, where it might be desirable to assign priorities to e-mail traffic through an MTA.
A final problem to store-and-forward processing is the lack of any mechanism for introducing any additions to, or modifications of, messages. Closely related to the problem of SMTP feed, the fact that a fixed message is input into the MTA and undergoes a rigidly defined set of processes for delivery to an end recipient, strictly limits the functionality of the MTA to relaying messages. With advances in e-mail processes such as secure certificates, customized content and the like, there are now many applications that make it highly desirable to modify messages in transit from the sender application (MDA or other system) to the Internet. Similar to the problem of using two or more systems to generate a message with reference to the SMTP feed, the inability to modify the content of messages traveling through an MTA leads to the current art in which applications that generate content modifications for e-mails must be integrated into the sending application for processing prior to sending.
Problems with Combined Validation and Transmission and Relay
It is commonly known that, in the current state of the art of MTA e-mail sending, once the MTA identifies the recipient server and opens a connection, a series of information transactions occurs in which the sending MTA validates the availability of the recipient server and the recipient server validates the deliverability of the message. In the current art, the MTA, irrespective of the results of validation, sends the message on to the recipient MTA. If the message ultimately proves to be undeliverable, the recipient server will send an SMTP error message back to the sending MTA, which will then resend the message. This process generally continues through several tries at which point the sending MTA will report a failure back to the originating sender or system, utilizing significant processing overhead, and bandwidth, for both original undeliverable e-mails and for the bounces and subsequent resends pursuant to them.
It is a longstanding problem in e-mail delivery to improve this process by validating the deliverability of a recipient e-mail address prior to sending a message and, in the event that it is not deliverable, not to send it. This problem has also been pursued for the purpose of validating deliverability for e-mail addresses in a large list to eliminate those that are permanently undeliverable from any future attempts to send e-mail. To date, attempts have focused on the use of a verification process built into sendmail for this purpose, the VRFY command, which was designed to operate independently of the standard SMTP communication process for sending e-mail. This standalone validation process could, however, also be used by operators external to a network to identify, via automated trial end error, valid e-mail addresses within the network, thereby posing a significant security hazard.
Referring now
The VRFY command is convenient since this feature is built into the commonly used application sendmail. However, since support of the VRFY process is an optional setting on recipient sendmail MTAs and is not critical to standard e-mail delivery operations, this feature is widely disabled as a security precaution. That is, rather than responding with an “Okay” as in step 112 or with a negative result, recipient MTA 40 responds to the VRFY command from the sender MTA with a notification that the VRFY feature has been disabled—thereby failing to confirm or deny the accuracy of the e-mail address of the intended recipient. Therefore, Applicants submit that the VRFY command on its own is not a solution to the problem of e-mail validation.
Problems with Application Program Interfaces for the MTA
As mentioned previously, growth in the use and applications of e-mail has led to the need and desire for processes that act on, modify, examine and otherwise interact with e-mail messages. Today, such processes include virus protection applications, SPAM filters, content filters, encryption, network signatures, policy filters, HTML template application and more. Moreover, while content-modifying applications can be applied through integration with initiating sender systems (albeit with difficulty) as mentioned previously, scanning processes are, by definition, applied to electronic messages as they travel into or across a network. An existing problem in the prior art electronic message delivery processes is the prevalence of an MTA architecture that combines current MTA processes with, at most, a single external process in one sequence. Furthermore, in cases where multiple external processes must be conducted on messages in a system or network, such external processes must be applied serially through a sequence of separate MTAs.
In the current state of the art such processes are applied serially, by sending each message through a separate application supported by a receiving MTA to accept messages and a separate, sending MTA to forward them on to the next processing application or the primary MTA for delivery through the Internet. As a consequence, the delivery efficiency of the entire system is limited by the delivery efficiency of the layers of serial processing through multiple MTAs and processing applications. The architecture is also hardware and support intensive.
Suggested Solutions
As described above, inherent limitations in the SMTP feed of currently available MTAs prohibit or limit the ability of the prior art MTAs to provide features such as, for example, increased delivery speed and combination of database information from multiple source systems. As a solution to these problems, Applicants submit an MTA that can accept data elements, raw data files or other electronic message formats though an application program interface (API), and use relevant data items to build SMTP e-mails within the MTA application. The electronic message may be, for example, an SMTP message or e-mail, or other formats of a package of electronic data intended for delivery from a sender to an intended recipient.
Additional problems exist in prior art MTAs due to the use of the store-and-forward mode of communication. Applicants submit that one solution to this problem is to store messages in Random Access Memory (RAM—i.e., electronic memory) that is accessible directly by the processor chip, thus eliminating the mechanical action associated with file I/O to realize a dramatic increase in efficiency. Early in the development of e-mail, the file I/O method was designed to ensure the recoverability of messages stored in process in the event of a system failure, and incorporated in the basic architecture of the modern MTA. Today, the risks of losing messages in process are drastically reduced by advances that have increased delivery-processing speeds such that storage time is reduced to nearly nil, and by the potential ease with which mechanisms can be designed that write files to the hard drive when the system fails.
Another problem arising from the store-and-forward processing of the prior art MTAs is the sequential, first-in-first-out (FIFO) nature of message processing, which often leads to processing bottlenecks and precludes any form of traffic prioritization or queue management. As a solution to this problem, Applicants submit the enabling of the MTA to review the domain composition across a group of messages or a recipient list and to order the processing of messages for delivery to distribute connection attempts to individual domains across the list. In this way, the likelihood of slowdowns and failures by recipient MTAs and potential bottlenecks created by exhausting thread connections with attempts to deliver to the same problematic domain are minimized. Furthermore, the present invention provides a flexible, customizable process to prioritize and manage message traffic.
The present invention also provides a solution to the deliverability validation problem by reengineering the standard SMTP communication process to enable an alternate process for messages that receive deliverability failure responses during the standard SMTP conversation. In this process, these messages are not sent on to the recipient MTA but are instead retried via additional SMTP delivery communication attempts until they are validated as deliverable and then sent, or reported as delivery failures to the originating sender or system and terminated.
The closed architecture problem is also mitigated by the present invention. The present invention provides an MTA application that enables the tight integration of corollary e-mail processes such as scanning and modification with the MTA application itself through the availability of open interfaces, at any of a number of stages in the standard SMTP process, to compatible applications designed for such processes. Such an MTA application may accept e-mails, or e-mail equivalent data files, for delivery and direct them, through a direct data interface, through these added processes prior to delivering them on to a recipient MTA. These and other features of the present invention are described in further detail immediately hereinafter.
Turning now to
Continuing to refer to
Attention is now directed to
Continuing to refer to
It is emphasized that the enhancement processes in sender MTA 320 are facilitated by the fact that sender MTA 320 is configured to accept data packets into RAM rather than being solely dependent on complete electronic messages that are saved as files onto hard drives. Since the final electronic message is generated from the various data packets, electronic message delivery system 300 is referred to as a “build-and-forward” system rather than the previously described “store-and-forward” system of the prior art. The build-and-forward system provides significant advantages over the store-and-forward system because the enhancement processes included in the build-and-forward system may be handled all in RAM rather than requiring a series of file reading and re-writing as in the store-and-forward system. Also, when deliverability confirmation is included in the enhancement processes, sender MTA 320 may be configured such that it does not generate final electronic message 350 if deliverability of that final electronic message 350 is not confirmed, thus resulting in savings in processing time and communication bandwidth.
In contrast,
Referring now to
Referring now to
Similarly,
In contrast,
The approach of the present invention in electronic message delivery is additionally discussed immediately hereinafter.
General Approach
The general approach to sending messages using this refactored architecture for SMTP message delivery includes the feeding of the server with messages for delivery, initiating the SMTP communication with the receiving server, building the message for delivery, and finally transferring the message content to the receiving server. We will address each in turn immediately hereinafter.
The server is first fed with sender and recipient information. This process can include the acceptance of a complete “well-formed” SMTP message, or a collection of data packets that will be combined in process to create the message. In the preferred implementation the input would be a collection of data packets/points for later message construction. This data would include information for the message headers, such as recipient e-mail address, subject lines, from address, etc. It would also include information for the message body. This could include text, HTML code, graphics, stored personal information from a separate data source, audio or video files, attachments or any of a number of other types of electronic data.
Once the system has the required data points for a valid SMTP connection, it initiates a connection with the receiving server. By opening a communication socket to the receiving server and beginning the normal SMTP communication, the system can determine the “deliverability” of the message. If deliverability is confirmed in the affirmative, the system builds a complete message formatted for the relevant protocol (e.g. SMTP) for delivery based on the data points provided. If deliverability is denied, the system flags the message as undeliverable and closes the connection to the receiving server.
Message delivery only occurs after a successful deliverability test, and message construction. Once these events have occurred, the message is transferred across the communication socket to the recipient server and the system closes the connection.
Description of Each of the Above Steps in Detail
If one were to follow this process in detail as it applies to an electronic message, it would progress as in the process outlined immediately hereinafter.
-
- a) A collection of data packets that contains all the information necessary to create the desired message is accepted by the system.
- b) The sending server examines the data packets for the domain name of the receiving server.
- c) The sending server uses this domain name to look up the mail exchange (MX) record using the Internet Domain Name Service (DNS) interface. This process returns the actual IP address of the receiving server for this message.
- d) The sending server establishes a TCP/IP socket connection to the IP address of the receiving server
- e) Upon connection, the receiving server sends an announcement of itself
- f) The sending server issues a “HELO” command with its' name as the argument.
- g) The receiving server then verifies that the sending server is valid and able to send e-mail into its system. If accepted, the receiving server will respond with a “250 Welcome” command. If the connection is refused, the reviving server will respond with a “500 connection refused” command.
- h) At this point the sending server will announce who is sending the e-mail with a “MAIL FROM:senderprefix@sernderdomain.com” command.
- i) Once again, the receiving server verifies that the sender's address is allowed to deliver e-mail into the receiving server. If the e-mail is acceptable, the receiving server responds with a “250 Okay” command. If the message is not acceptable, the receiving server responds with a “500 Not allowed” command.
- j) At this point the sending server sends the “RCPT TO:recipientprefix@recipientdomain” command to tell the receiving server who is to receive the message.
- k) The receiving server verifies the recipient address, checking that it is an active address within the domain and available for receiving a message. If the recipient address is valid and available, the receiving server will respond with a “250 Okay” command. If the recipient address is not valid, the receiving server will respond with a “500 User unknown” or other user availability failure command.
- l) It is at this point in the process where the invention departs from the current state of the art. Once the verification transactions in the SMTP conversation have occurred, the sending server will, in the case of positive verification of deliverability, build the message and continue the send process, or, in the case of a deliverability failure error, write the failure to a record log, discontinue the process and close the connection.
- m) In this new process, once a message is built, the sending server then sends the “DATA” command.
- n) If the receiving server is ready to accept the message, it will send a command in the format “250 Okay”
- o) Once the sending server has the aforementioned okay, it will send the full message content to the receiving server, ending with a “<cr><lf><cr><lf>” character sequence. This string will inform the receiving server that the message is complete.
- p) If the receiving server accepts the message it will send another “250 Okay” command.
- q) Once the final okay is received, the sending server issues a “QUIT” command, and upon receipt the receiving server will disconnect the socket and both side will close communication.
Optional Additional Features
Build-and-Forward (Data Packet-Based)
Build-and-forward messaging is comprised of three major steps. The first, and most important, is the ability to accept data packets containing recipient, delivery and message content information and store them in Random Access Memory (RAM), rather than writing them to a hard drive, while an SMTP communication is established to send a group of messages. The second key feature is to hold a single copy of the main e-mail content (body and certain headers), while processing large volumes of unique recipient data packets containing delivery and other personalized information. The third step in a “build and forward” process is to merge the e-mail content with each recipient's unique delivery and other personalized information, after the SMTP connection is established and deliverability has been determined and before (or during) transmission to the receiving server.
Message Handles in Memory (Written to RAM, Not Disk)
A key feature of this system is the handling of messages in electronic memory. This is done with addressable RAM resources, rather than file I/O. By using this resource, rather then the typical hard drive of a system, the bottleneck associated with file I/O as well as the single point of failure common to the prior art may be reduced.
Dynamic IP Address Allocation
Dynamic IP address allocation is employed to spread the delivery load across a much wider address space. This allows for faster delivery and reduces the likelihood that the messages being delivered at volume will be seen as unsolicited commercial mail.
Serial Approach (Handshake/Send for each Recipient) Vs. Parallel Approach (One Confirmation Connection, A Different Delivery Connection)
This deliverability confirmation step can be taken in a serial fashion, i.e. one step after another utilizing the same socket connection, or in parallel to the actual delivery. Both approaches have merit, and result in the same conclusion.
Sender May be Human or Application/Machine
Since the primary acceptance method for this system is Data, rather then well-formed SMTP messages, the originator may be either human or a computer system. The use of this Data driven implementation does not preclude the use of SMTP messages, but rather, extends the compatibility.
Data Driven Support
By breaking down the message transfer into distinct data packets, the system can interface with existing and future applications, as well as expand the possibilities of message enhancement during the message delivery process, as opposed to before delivery.
-
- a) API connection with various software, features: By establishing an API “chain”, each message can be handed to a modifier application. Such applications might serve functions for virus protection, SPAM filtering, content management, etc.
- b) Triggered e-mail: As messages are composed of data packets in the proposed system, this system lends itself to the modification and delivery of e-mail that is driven by database triggers. Database triggers occur when specified criteria are met in the database. The database can the “trigger” an event such as the delivery of an e-mail. By exposing the message process to data packets are building messages late in the delivery process, triggered messaging becomes practical.
- c) Monitoring of outbound messages and events: In addition to the added functionality that data driven messaging can provide, this data approach allows the administrator of the system to monitor and control all messaging activities within the system. From a simple count of the number of messages, to advanced API chain management.
Domain-Based Flow Control
Domain-based flow control is achieved by looking at the complete group of e-mail domains that are scheduled for delivery and “normalizing” the distribution so that no single domain can block delivery of the other domains. As is well known in the art, any list of e-mail addresses is composed of a majority of the large ISP's, then filled out with the rest of the smaller service providers. In a typical list sending scenario, the e-mails are delivered in groups based on domain name. If one or more of the large ISP's has difficulty accepting mail or slows down, the remaining portion of the list will wait for the pervious mail to be sent. (classic FIFO behavior) This behavior clearly limits a systems ability to deliver large volumes of mail quickly. Domain based flow control is achieved by dividing each domain into equal size portions and delivering based on these groups rather then a strict domain ordering.
In-Bound Message Reception
As well as the data driven nature of the system, it will also allow messages to be into the system using a standard SMTP protocol. This allows expanded functionality as described further.
Data Feedback into the Database, According to Handshake/Delivery Results
A powerful feature of this system is the ability to log bounces and failures in the mail delivery without actually sending the e-mail. This is accomplished by logging the response from the receiving server before message creation and delivery.
Data Driven Support
Another aspect of the system is the ability to handle one or more data sources as input for each message. This capability frees the standard mail server from the previous art notion of “mail relay”, instead creating a message factory within the mail delivery system.
Macro Replacement Fields, Enhanced E-Mail, Personalization, Modification of Content
The advancement mentioned above, enables each message to contain personalized data pulled from disparate data sources and built into the message via custom business logic, or standard macro replacement. This provides infinite flexibility in the message creation process and is in stark contrast to the previous art.
Architecture Flexibility
A key feature if the proposed system is the ability to extend and expand the mail system architecture. By allowing an “in-line” API to messages traveling within the system, messages can be enhanced, modified or manipulated without the serial approach common in the previous art. This process acts as follows:
-
- i. A message is accepted into the system
- ii. The initial message related processing occurs
- iii. The message is handed to “queue manager” which will allow ordered processing of the API chain
- iv. The message is then handed to each API process registered within the system
- v. Once the message has traversed the API chain it continues its delivery path.
- vi. Then the message is then delivered to the final recipient.
Other Message Delivery Protocols
In addition to the e-mail messaging protocol discussed to this point, it should be noted that this system can be used to enable all of the mentioned features for other current, or future internet protocols. This includes SMS (Simple Messaging Service), and various other protocols that are currently available or may be developed in the future.
Adaptation for Use as In-Bound/Out-Bound MTA
In addition to expanded protocols, this system can be turned to face the inside of a network to allow inbound message handling with the same advances in the art. As a feature of inbound processing or message relay, the system could be configured to attempt retries on messages where deliverability can be wholly determined, or a message is clearly instructed to re-send at a later time. This system can also be expanded to act as a true messaging relay to increase the performance and capabilities of the pervious art.
Additional Modifications
API Event Management and Distribution
One example of prior art API process is the aforementioned MILTER process. Referring to
While the aforedescribed MILTER architecture enables interfacing of the MTA with external applications for various types of processing, the limitations of MTA performance, architecture and features have limited the nature of MILTER architecture applicability. Specifically, the MILTER interface is commonly used to interface a single e-mail processing application (e.g., a virus filter) with an MTA for the purpose of leveraging the MTA as a mail relay mechanism for the application. Where more than one such application (e.g., a virus filter and a spam filter) is required for a network or system, multiple MTAs are deployed in sequence, feeding messages one to another as shown in
The prevalence of this problematic, sequential processing scheme is largely due to MTA limitations in the current art that include slow processing speed, inefficient use of hardware resources, inefficient data traffic management, lack of efficient queuing methods for messages awaiting processing in the context of multiple available processes and the lack of a means for ordering messages as they travel through multiple processes. These limitations combine to create unacceptable risks of overuse of available system resources, traffic bottlenecks and potential system failure when multiple API processes are implemented in a single instance of an MTA. Therefore, in the current state of the art, such processes are almost always applied serially, by sending each message through a separate application supported by a receiving MTA to accept messages and a separate, sending MTA to forward them on to the next processing application or the primary MTA for delivery through the Internet (as exemplified, for example, by delivery engine 800 in
As an alternative to the API feed process of the present invention as described in conjunction with
It is known that the delivery process of an electronic message using SMTP protocol includes a plurality of message events such as, for example, Connect, HELO, MAIL FROM, RCPT TO, DATA and QUIT. In the API process of the present invention, the API application may register for these message events anytime during the delivery process. In other words, registration may occur at instances Post_Connect, Post_HELO, Post_MAIL, Post_RCPT, Post_Data, Post_QUIT, Msg_Injector, Pre_Connect, Pre_HELO, Pre_MAIL, Pre_RCPT, Pre_DATA and Pre_QUIT.
Examples of Post- and Pre-event registration are illustrated in
The addition of events that may be triggered before a messaging event (e.g., pre_HELO) allows filter applications to be applied to messages on their way OUTBOUND from a server, which is considered to be a significant advancement over the prior art MILTER process. In the API process of the present invention, an application may register with the mail server over a local or remote TCP socket connection. This registration of the application specifies the event that the sender would like to capture and the order in which that particular sender should be positioned in the listener queue. This API event registration concept of the present invention is illustrated in
Referring now to
Configurable RAM or Hard Drive Message Processing
As an example of a possible modification to the aforedescribed embodiments, the electronic message delivery system of the present invention may be configured to provide a choice between handling the messages as data packets in RAM processing, as described above, and as files saved on a hard disk drive (HDD), as used in the prior art. That is, as a message, which is intended for delivery to a recipient at a recipient server, is received from a sender, the sending mail server may be configured to write a part or all of this message to HDD for storage. This option of storing part or all of the message to HDD complements the RAM-based processing of the present invention as described above by providing, for example, a backup storage system in case of a power outage during the RAM processing. Such a modified system is illustrated in
As still another alternative, the configurable RAM/HDD message processing may be modified to determine whether or not a given message should be saved to HDD based on a predetermined system threshold. Such a threshold-based process within the mail server is similar to the process illustrated in
Channel Selector Socket Management
Another consideration in the mail server art is the use of socket communication for email transmission between servers. As a brief background, a socket is a mechanism for creating a physical remote or local connection between processes. Sockets interface the standard input/output (I/O) function stream services of an operating system (e.g., UNIX) via its network communication facilities. The socket function creates a communications end-point and returns a file descriptor with which to access that socket. The socket has associated with it a socket address, consisting of a port number and the local host's network address.
The classic, socket communication model is a single socket model illustrated in
Referring now to
-
- a. In a step 1112, create a socket at the sending server (origin).
- b. In a step 1114, connect the socket from the sending server to the receiving server (end point).
- c. In a step 1116, the recipient server elects to accept the socket connection. Once the connection is established and accepted, the socket communications channel is available to transfer information, including both standard electronic message protocol conversation data and message content, between the origin and the end point.
- d. In a step 1118, the sending server sends data to and receives data from the receiving server.
- e. The receiving server sends and receives data in a step 1120.
- f. Following step 1120, a decision 1122 is made to determine whether or not more data need to be transferred between the sending and receiving servers. If the answer to decision 1122 is YES, more data need to be exchanged between the sending and receiving servers, then the process cycles back through steps 1108 and 1110. If the answer to decision 1122 is NO, the flow of data is complete, then the sending server terminates the connection in a step 1124.
Both sides (i.e., origin and end point) remain active while connected through the socket. Either side may interrupt or close the connection. The connection may be used to transmit multiple e-mail messages (from a waiting queue) to the end point server. Upon completion of the transmission, interruption or closure by either one of the servers, the connection is severed and the socket disappears. The sequence from socket creation to information transfer to closing of the connection and socket disappearance is standard and uniform in the current state of the art of e-mail transmission between servers. Generally, open sockets will only exist where there has been a continuous flow of messages/data to the end point location (i.e., a new message to the location is presented to the socket prior to the completion of transmission of the current message).
There are various problems inherent in the classic, single socket communication process of the prior art. In the prior art process illustrated in
Recent advances in data communication between servers, illustrated in
First referring to
Referring now to
The prior art socket channels and selector TCP process as shown in
In order to circumvent the aforedescribed problems of prior art and adapt a socket communication process for use in e-mail delivery, Applicants have devised a number of data communication methods which provide alternatives to the socket communication method of the prior art. In particular, Applicants have developed an improved electronic messaging server based on the channel and selector TCP method but providing additional advantages over the socket channels and selector scheme as illustrated in
Referring first to
Referring to
The channel and selector TCP method described above may be used to send and receive email via SMTP and other electronic messaging protocols. While the selector is waiting for a response from the End Point for a given socket channel, the selector is free to initiate, manage, communicate with or close other socket connections. This method is advantageous because it greatly increases the throughput capabilities for SMTP data traffic, and thus the performance of email delivery servers, by dramatically increasing the efficiency of data transmissions.
Prior art e-mail delivery technologies, before the innovations developed by Applicants, yield 32 messages/second peak throughput. With the inclusion of the aforedescribed innovations by Applicants, but without the socket channel and selector TCP process of the present invention, the message delivery system achieved 150 messages/second peak throughput. Prior to actual implementation of the aforedescribed socket channel and selector TCP process of the present invention, the expected performance improvement for SMTP email delivery was approximately 50 to 75% improvement over the 150 messages/second peak throughput (for 10,000-byte messages) already achieved. Upon implementation, however, it has been found that the socket channel and selector TCP process of the present invention, in combination with the earlier described improvements developed by Applicants, yields a message throughput of over 800 messages/second (for 10,000 messages), resulting in an extraordinary 530% increase in performance achieved by Applicants without the socket channel and selector TCP process and 2500% over the known prior art. Applicants submit this improvement is significant and unexpected.
The channel and selector method described in reference to
The channel selector socket management process using UDP may be used, for example, in a DNS resolver to greatly improve the speed of DNS lookup processes. The DNS resolver of the present invention utilizes the socket channel and selector process to send and request domain name lookup data using standard DNS protocol. Such a DNS resolver may be included within the originating mail server itself so as to facilitate the transmission and reception of e-mail. The channel selector socket management processes using TCP or UDP may be combined with, for example, the deliverability confirmation process described above.
Prior to actual implementation of the DNS resolver of the present invention, the anticipated performance improvement for DNS resolution (i.e., lookups) using the socket channel and selector process was approximately a 100% improvement over the current state of the art method (BIND), which has an average peak performance rate of 50 lookups per second. Unexpectedly, Applicants have found that the DNS resolver based on the socket channel and selector process of the present invention yields an astonishing rate of 3400 lookups per second, thus providing a 6800% increase in performance over the existing prior art.
The socket channel and selector process may be combined, for example, with the aforedescribed API process of the present invention in order to achieve a higher performance API application with very low latency on message transmission from the server to the API application.
Although each of the aforedescribed embodiments have been illustrated with various components having particular respective orientations, it should be understood that the present invention may take on a variety of specific configurations with the various components being located in a wide variety of positions and mutual orientations and still remain within the spirit and scope of the present invention. Furthermore, suitable equivalents may be used in place of or in addition to the various components, the function and use of such substitute or additional components being held to be familiar to those skilled in the art and are therefore regarded as falling within the scope of the present invention. For example, although each of the aforedescribed embodiments have been discussed mainly using the case of e-mail messaging, the electronic message delivery system of the present invention may be readily configured to accommodate other types of electronic messages such as, for instance, messages to and from cellular phones and other handheld devices.
Therefore, the present examples are to be considered as illustrative and not restrictive, and the invention is not to be limited to the details given herein but may be modified within the scope of the appended claims.
Claims
1. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said message including at least recipient information and message body, said method comprising:
- testing said recipient information by establishing a temporary test connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, so as to provide an indication of deliverability of said message to said recipient without using an SMTP VRFY command.
2. The method of claim 1 further comprising:
- based on said indication, performing one of (i) establishing a delivery connection through the Internet from said sender server to said recipient server then directing said message to said recipient server for delivery to said recipient, and (ii) aborting the delivery of said message to said recipient.
3. The method of claim 2 wherein said message further includes sender information, said method further comprising:
- before establishing said delivery connection, selecting a plurality of IP addresses at said sender server;
- specifying a certain one of said plurality of IP addresses according to a set of predetermined rules; and
- including said certain one of said plurality of IP addresses as a part of said sender information of said message.
4. The method of claim 1 wherein testing said recipient information includes querying the recipient server to verify that the recipient information is valid.
5. The method of claim 1 wherein testing said recipient information includes verifying the availability of the recipient server to receive an incoming message.
6. The method of claim 1 further comprising:
- creating a record regarding the deliverability of said message.
7. The method of claim 1 wherein said messaging system includes an electronic memory for processing data and a hard drive for storing files, and said method further comprising:
- handling said message as data in said electronic memory without storing said message as one of said files in said hard drive.
8. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said message including at least recipient information and message body, said method comprising:
- testing said recipient information by establishing a connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, in a way which provides an indication of deliverability to said recipient without using an SMTP VRFY command; and
- based on said indication, performing one of (i) directing said message to said recipient server through said connection to said recipient, and (ii) aborting the delivery of said message.
9. The method of claim 8 wherein said message further includes sender information, said method further comprising:
- before directing said message to said recipient server, selecting a plurality of IP addresses at said sender server;
- specifying a certain one of said plurality of IP addresses according to a set of predetermined rules; and
- including said certain one of said plurality of IP addresses as a part of said sender information of said message.
10. The method of claim 8 wherein testing said recipient information includes querying the recipient server to verify that the recipient information is valid.
11. The method of claim 8 wherein testing said recipient information includes verifying the availability of the recipient server to receive an incoming message.
12. The method of claim 8 further comprising:
- creating a record regarding the deliverability of said message.
13. The method of claim 8 wherein said messaging system includes an electronic memory for processing data and a hard drive for storing files, and said method further comprising:
- handling said message as data in said electronic memory without storing said message as one of said files in said hard drive.
14. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said message including at least recipient information and message body, said method comprising:
- confirming deliverability of said message to said recipient according to said recipient information, wherein confirming includes
- establishing a connection through the Internet between said sender server and said recipient server, and
- from said sender server, using said connection to verify said recipient information with said recipient server so as to provide an indication, without using an SMTP VRFY command, as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server.
15. The method of claim 14 further comprising:
- based on said indication, performing one of (i) directing said message so generated to said recipient server through said connection for delivery to said recipient, responsive to said positive confirmation, and (ii) aborting the delivery of said message, responsive to said renouncement.
16. The method of claim 14 wherein said message further includes sender information, said method further comprising:
- before directing said message to said recipient server, selecting a plurality of IP addresses at said sender server;
- specifying a certain one of said plurality of IP addresses according to a set of predetermined rules; and
- including said certain one of said plurality of IP addresses as a part of said sender information of said message.
17. The method of claim 14 wherein using said connection includes querying the recipient server to verify that the recipient information is valid.
18. The method of claim 14 wherein using said connection includes ensuring the availability of the recipient server to receive an incoming message.
19. The method of claim 14 further comprising:
- creating a record regarding said indication.
20. The method of claim 14 wherein said messaging system includes an electronic memory for processing data and a hard drive for storing files, and said method further comprising:
- handling said message as data in said electronic memory without storing said message as one of said files in said hard drive.
21. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said message including at least recipient information and message body, said method comprising:
- confirming deliverability of said message to said recipient according to said recipient information, wherein confirming includes
- establishing a temporary test connection through the Internet between said sender server and said recipient server, and
- from said sender server, verifying said recipient information with said recipient server through said temporary test connection so as to provide an indication, without using an SMTP VRFY command, as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server.
22. The method of claim 21 further comprising:
- based on said indication, performing one of (i) generating said message by combining at least said recipient information and message body, responsive to said positive confirmation, then establishing a delivery connection through the Internet between said sender server and said recipient server followed by directing said message so generated to said recipient server through said delivery connection for delivery to said recipient, and (ii) aborting the delivery of said message, responsive to said renouncement.
23. The method of claim 21 wherein using said temporary test connection includes querying the recipient server to authenticate that the recipient information is valid.
24. The method of claim 21 wherein using said temporary test connection includes ensuring the availability of the recipient server to receive an incoming message.
25. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said method comprising:
- at said sender server, receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- testing said recipient information by establishing a temporary test connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, so as to provide an indication of deliverability to said recipient without using an SMTP VRFY command; and
- based on said indication, performing one of (i) generating said message by combining at least said recipient information and message body then establishing a delivery connection through the Internet from said sender server to said recipient server followed by directing said message through said delivery connection to said recipient server for delivery to said recipient, and (ii) aborting the delivery of said message.
26. The method of claim 25 wherein said message further includes sender information, and wherein generating said message includes
- selecting a plurality of IP addresses at said sender server,
- specifying a certain one of said plurality of IP addresses according to a set of predetermined rules; and
- including said certain one of said plurality of IP addresses as a part of said sender information of said message.
27. The method of claim 25 wherein receiving said plurality of discrete data packets from said sender includes processing said plurality of discrete data packets as created by a human user.
28. The method of claim 25 wherein receiving said plurality of discrete data packets from said sender includes processing said plurality of discrete data packets as created by a software application.
29. The method of claim 28 further comprising:
- interfacing said software application with a database;
- defining a trigger event in said database such that, when said trigger event occurs in said database, said software application automatically generates said plurality of discrete data packets.
30. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said method comprising:
- at said sender server, receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- testing said recipient information by establishing a connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, in a way which provides an indication of deliverability to said recipient without using an SMTP VRFY command; and
- based on said indication, performing one of (i) generating said message by combining at least said recipient information and message body then directing said message through said connection to said recipient server for delivery to said recipient, and (ii) aborting the delivery of said message.
31. The method of claim 30 wherein receiving said plurality of discrete data packets from said sender includes processing said plurality of discrete data packets as created by a human user.
32. The method of claim 30 wherein receiving said plurality of discrete data packets from said sender includes processing said plurality of discrete data packets as created by a software application.
33. The method of claim 32 further comprising:
- interfacing said software application with a database;
- defining a trigger event in said database such that, when said trigger event occurs in said database, said software application automatically generates said plurality of discrete data packets.
34. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said method comprising:
- at said sender server, receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- confirming deliverability of said message to said recipient according to said recipient information by establishing a connection through the Internet between said sender server and said recipient server, and
- from said sender server, verifying said recipient information with said recipient server through said connection so as to provide an indication, without using an SMTP VRFY command, as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server; and
- based on said indication, performing one of (i) generating said message by combining at least said recipient information and message body, responsive to said positive confirmation, then directing said message so generated to said recipient server through said connection for delivery to said recipient, and (ii) aborting the delivery of said message, responsive to said renouncement.
35. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said method comprising:
- at said sender server, receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- confirming that said message is deliverable to said recipient according to said recipient information by establishing a temporary test connection through the Internet between said sender server and said recipient server, and
- from said sender server, verifying said recipient information with said recipient server through said temporary test connection so as to provide an indication, without using an SMTP VRFY command, as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server; and
- based on said indication, performing one of (i) generating said message by combining at least said recipient information and message body, responsive to said positive confirmation, then establishing a delivery connection through the Internet between said sender server and said recipient server followed by directing said message so generated to said recipient server through said delivery connection for delivery to said recipient, and (ii) aborting the delivery of said message, responsive to said renouncement.
36. A modified message server configured to replace an unmodified message server interfaced to the Internet for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through said unmodified message server capable of communicating with a recipient server, said unmodified message server including a hard drive configured for saving said message as a message file thereon, and said unmodified message server being configured to forward said message file to said recipient server, said modified message server comprising:
- means for receiving from said sender as a plurality discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body, said modified receiving arrangement including electronic memory for saving said plurality of discrete data packets therein; and
- means for processing said plurality of discrete data packets and generating said message by combining at least said recipient information and message body.
37. The modified message server of claim 36 further comprising:
- means for testing said recipient information by establishing a connection through the Internet from said modified message server to said recipient server and using said connection to verify said recipient information with said recipient server, before relying on said recipient information for sending said message, so as to provide an indication, without using an SMTP VRFY command, as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server.
38. The modified message server of claim 37 wherein said processing means is configured to perform, based on said indication, one of (i) directing said message so generated to said recipient server through said connection for delivery to said recipient, responsive to said positive confirmation, and (ii) aborting the delivery of said message, responsive to said renouncement.
39. The modified message server of claim 37 wherein said processing means is configured to generate said message only after said testing means provides said positive confirmation.
40. In a messaging system interfaced through the Internet, a method for processing a group of messages from a sender, which sender is interfaced to the Internet through a sender server capable of communicating with a plurality of recipient servers, each one of said group of messages being intended for delivery to a recipient, said method comprising:
- at said sender server, receiving from said sender a plurality of discrete data packets for use in defining said group of messages, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body for a corresponding one of said group of messages;
- generating said group of messages by combining at least said recipient information and message body for each one of said group of messages; and
- arranging said group of messages in an order in accordance with a predetermined set of rules based on at least one of said recipient information and said message body of each one of said group of messages.
41. The method of claim 40 further comprising, for each one of said group of messages:
- testing said recipient information by establishing a connection through the Internet from said sender server to a corresponding one of said plurality of recipient servers, before relying on said recipient information for sending said one of said group of messages, in a way which provides an indication of deliverability to the corresponding one of said plurality of recipients without using an SMTP VRFY command; and
- based on said indication, performing one of (i) and directing said one of said group of messages through said connection for delivery through said corresponding one of said plurality of recipient servers to one of said plurality of recipients, and (ii) aborting the delivery of said one of said group of messages.
42. The method of claim 40 further comprising, for each one of said group of messages:
- testing said recipient information by establishing a temporary test connection through the Internet from said sender server to a corresponding one of said plurality of recipient servers, before relying on said recipient information for sending said one of said group of messages, in a way which provides an indication of deliverability to the corresponding one of said plurality of recipients without using an SMTP VRFY command; and
- based on said indication, performing one of (i) establishing a delivery connection through the Internet from said sender server to a corresponding one of said plurality of recipient servers, and directing said one of said group of messages through said delivery connection for delivery through said corresponding one of said plurality of recipient servers to one of said plurality of recipients, and (ii) aborting the delivery of said one of said group of messages.
43. The method of claim 40 wherein still another one of said plurality of discrete data packets includes sender information for a corresponding one of said group of messages, and wherein generating said group of messages further includes
- selecting a plurality of IP addresses at said sender server,
- specifying a certain one of said plurality of IP addresses according to a set of predetermined rules; and
- including said certain one of said plurality of IP addresses as a part of said sender information of each one of said group of messages.
44. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said method comprising:
- at said sender server, receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- directing said plurality of discrete data packets through an application program interface to one of a plurality of external applications for performing at least one of (i) modifying said plurality of discrete data packets according to a first predetermined set of rules, (ii) adding additional data packets to said plurality of discrete data packets according to a second predetermined set of rules, and (iii) ordering said plurality of discrete data packets according to a third predetermined set of rules, and, consequently, for providing a plurality of resultant data packets;
- receiving said plurality of resultant data packets from said one of a plurality of external applications through said application program interface;
- generating said message by combining at least a portion of said plurality of resultant data packets;
- establishing a connection through the Internet from said sender server to said recipient server; and
- directing said message through said connection to said recipient server for delivery to said recipient.
45. A messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said message including at least recipient information and message body, said messaging system comprising:
- a testing arrangement for testing said recipient information by establishing a temporary test connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, so as to provide an indication of deliverability of said message to said recipient without using an SMTP VRFY command.
46. The messaging system of claim 45 further comprising:
- a selective delivery arrangement for performing, based on said indication, one of (i) establishing a delivery connection through the Internet from said sender server to said recipient server then directing said message to said recipient server for delivery to said recipient, and (ii) aborting the delivery of said message to said recipient.
47. The messaging system of claim 45 further comprising an electronic memory for processing data and a hard drive for storing files, wherein said messaging system is configured to process said message as data in said electronic memory without storing said message as one of said files in said hard drive.
48. A messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said message including at least recipient information and message body, said messaging system comprising:
- a testing arrangement for testing said recipient information by establishing a connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, in a way which provides an indication of deliverability to said recipient without using an SMTP VRFY command; and
- a selective delivery arrangement for performing, based on said indication, one of (i) directing said message to said recipient server through said connection for to said recipient, and (ii) aborting the delivery of said message.
49. A messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server and which recipient is interfaced to the Internet through a recipient server, said message including at least sender information, recipient information and message body, messaging system comprising:
- a testing arrangement for confirming deliverability of said message to said recipient according to said recipient information by
- establishing a connection through the Internet between said sender server and said recipient server, and
- from said sender server, using said connection to verify said recipient information with said recipient server so as to provide an indication without using an SMTP VRFY command as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server.
50. The messaging system of claim 49 further comprising a selective delivery arrangement for performing, based on said indication, one of (i) directing said message so generated to said recipient server through said connection for delivery to said recipient, responsive to said positive confirmation, and (ii) aborting the delivery of said message, responsive to said renouncement.
51. A messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said message including at least recipient information and message body, said messaging system comprising:
- a testing arrangement for confirming deliverability of said message to said recipient according to said recipient information by
- establishing a temporary test connection through the Internet between said sender server and said recipient server, and
- from said sender server, verifying said recipient information with said recipient server through said temporary test connection so as to provide an indication, without using an SMTP VRFY command, as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server.
52. The messaging system of claim 51 further comprising a selective delivery arrangement for performing, based on said indication, one of (i) generating said message by combining at least said recipient information and message body, responsive to said positive confirmation, then establishing a delivery connection through the Internet between said sender server and said recipient server followed by directing said message so generated to said recipient server through said delivery connection for delivery to said recipient, and (ii) aborting the delivery of said message, responsive to said renouncement.
53. A messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said messaging system comprising:
- at said sender server, a receiving arrangement for receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- a testing arrangement for testing said recipient information by establishing a temporary test connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, so as to provide an indication of deliverability to said recipient without using an SMTP VRFY command; and
- a selective delivery arrangement for performing, based on said indication, one of (i) generating said message by combining at least said recipient information and message body then establishing a delivery connection through the Internet from said sender server to said recipient server followed by directing said message through said delivery connection to said recipient server for delivery to said recipient, and (ii) aborting the delivery of said message.
54. A messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said messaging system comprising:
- at said sender server, a receiving arrangement for receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- a testing arrangement for testing said recipient information by establishing a connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, in a way which provides an indication of deliverability to said recipient without using an SMTP VRFY command; and
- a selective delivery arrangement for performing, based on said indication, one of (i) generating said message by combining at least said recipient information and message body then directing said message through said connection to said recipient server for delivery to said recipient, and (ii) aborting the delivery of said message.
55. A messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said messaging system comprising:
- at said sender server, a receiving arrangement for receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- a testing arrangement for confirming deliverability of said message to said recipient according to said recipient information by
- establishing a connection through the Internet between said sender server and said recipient server, and
- from said sender server, verifying said recipient information with said recipient server through said connection so as to provide an indication, without using an SMTP VRFY command, as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server; and
- a selective delivery arrangement for performing, based on said indication, one of (i) generating said message by combining at least said recipient information and message body, responsive to said positive confirmation, then directing said message so generated to said recipient server through said connection for delivery to said recipient, and (ii) aborting the delivery of said message, responsive to said renouncement.
56. A messaging system, interfaced through the Internet, for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said messaging system comprising:
- at said sender server, a receiving arrangement for receiving from said sender a plurality of discrete data packets for use in defining said message, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- a testing arrangement for confirming that said message is deliverable to said recipient according to said recipient information by
- establishing a temporary test connection through the Internet between said sender server and said recipient server, and
- from said sender server, verifying said recipient information with said recipient server through said temporary test connection so as to provide an indication, without using an SMTP VRFY command, as to whether said recipient is interfaced to the Internet through said recipient server, said indication being one of a positive confirmation that said recipient is indeed interfaced to the Internet through said recipient server and a renouncement showing that said recipient is not interfaced to the Internet through said recipient server; and
- a selective delivery arrangement for performing, based on said indication, one of (i) generating said message by combining at least said recipient information and message body, responsive to said positive confirmation, then establishing a delivery connection through the Internet between said sender server and said recipient server followed by directing said message so generated to said recipient server through said delivery connection for delivery to said recipient, and (ii) aborting the delivery of said message, responsive to said renouncement.
57. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said messaging system including an electronic memory for processing data, said method comprising:
- handling said message as data in said electronic memory.
58. The method of claim 57 wherein said messaging system further includes a hard drive for storing files, and wherein said handling said message includes handling said message without storing said message as one of said files in said hard drive.
59. The method of claim 57 further comprising:
- at said sender server, separating said message into a plurality of discrete data packets, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body; and
- directing said plurality of discrete data packets through an application program interface to one of a plurality of external applications for performing at least one of (i) modifying said plurality of discrete data packets according to a first predetermined set of rules, (ii) adding additional data packets to said plurality of discrete data packets according to a second predetermined set of rules, and (iii) ordering said plurality of discrete data packets according to a third predetermined set of rules, and, consequently, for providing a plurality of resultant data packets.
60. The method of claim 59 further comprising:
- receiving said plurality of resultant data packets from said one of a plurality of external applications through said application program interface;
- generating said message by combining at least a portion of said plurality of resultant data packets;
- establishing a connection through the Internet from said sender server to said recipient server; and
- directing said message through said connection to said recipient server for delivery to said recipient.
61. In a messaging system interfaced through the Internet, a method for processing a message from a sender, which sender is interfaced to the Internet through a sender server capable of communicating with a plurality of recipient servers, said message being intended for delivery to a recipient, said method comprising:
- at said sender server, separating said message into a plurality of discrete data packets, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- testing said recipient information by establishing a connection through the Internet from said sender server to a recipient server, before relying on said recipient information for sending said message, in a way which provides an indication of deliverability to the recipient without using an SMTP VRFY command; and
- based on said indication, performing one of (i) directing said message through said connection for delivery through said recipient server to said recipient, and (ii) aborting the delivery of said message.
62. In a messaging system interfaced through the Internet, a method for processing a message from a sender, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said message being intended for delivery to a recipient, said method comprising:
- at said sender server, separating said message into a plurality of discrete data packets, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body;
- testing said recipient information by establishing a temporary test connection through the Internet from said sender server to a recipient server, before relying on said recipient information for sending said message, in a way which provides an indication of deliverability to the recipient without using an SMTP VRFY command; and
- based on said indication, performing one of (i) establishing a delivery connection through the Internet from said sender server to said recipient server, and directing said message through said connection for delivery through said recipient server to said recipient, and (ii) aborting the delivery of said message.
63. In a messaging system interfaced through the Internet, a method for processing a group of messages from a sender, which sender is interfaced to the Internet through a sender server capable of communicating with a plurality of recipient servers, each message in said group of messages being intended for delivery to a recipient, said method comprising:
- at said sender server, separating each message in said group of messages into a plurality of discrete data packets, at least one of said plurality of discrete data packets including recipient information and at least another one of said plurality of discrete data packets including message body for the corresponding message out of said group of messages;
- arranging said group of messages in an order in accordance with a predetermined set of rules based on at least one of said recipient information and said message body for each message out of said group of messages.
64. The method of claim 63 further comprising:
- directing said plurality of discrete data packets through an application program interface to one of a plurality of external applications for performing at least one of (i) modifying said plurality of discrete data packets according to a first predetermined set of rules, (ii) adding additional data packets to said plurality of discrete data packets according to a second predetermined set of rules, and (iii) ordering said plurality of discrete data packets according to a third predetermined set of rules, and, consequently, for providing a plurality of resultant data packets.
65. In a messaging system interfaced through the Internet, a method for processing a message from a sender and intended for delivery to at least one recipient, which sender is interfaced to the Internet through a sender server capable of communicating with a recipient server, said messaging system including an electronic memory for processing data and a hard drive for storing files, said method comprising:
- selecting at least one of (i) handling said message as data in said electronic memory without storing said message as one of said files in said hard drive, and (ii) storing said message as one of said files in said hard drive.
66. The method of claim 65 wherein said message includes at least recipient information and message body, said method further comprising:
- testing said recipient information by establishing a temporary test connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, so as to provide an indication of deliverability of said message to said recipient without using an SMTP VRFY command.
67. The method of claim 66 further comprising:
- based on said indication, performing one of (i) establishing a delivery connection through the Internet from said sender server to said recipient server then directing said message to said recipient server for delivery to said recipient, and (ii) aborting the delivery of said message to said recipient.
68. The method of claim 65 wherein said message includes at least recipient information and message body, said method further comprising:
- testing said recipient information by establishing a connection through the Internet from said sender server to said recipient server, before relying on said recipient information for sending said message, in a way which provides an indication of deliverability to said recipient without using an SMTP VRFY command; and based on said indication, performing one of (i) directing said message to said recipient server through said connection to said recipient, and (ii) aborting the delivery of said message.
Type: Application
Filed: Jun 25, 2004
Publication Date: Jan 13, 2005
Inventors: Scott Brown (Superior, CO), Kelly Wanser (Thornton, CO), Victor Ganora (Louisville, CO)
Application Number: 10/876,964