Abstract: Popular object requests are handled using a front-end cache, and hashing is applied only to the requests in the stream that were not handled by the front-end cache. A cache (e.g., a web proxy cache) is placed in front of a Level 7 switch and services the popular requests from the cache based on the content of the request (e.g., based on the portion of an HTTP request following the domain name). The remaining requests are hashed and then routed to the back-end server. Requests that make it past the cache are still routed to the back-end server. A Level 4 switch is placed in front of a plurality of web proxy caches, each of which are in turn placed in front of a respective Level 7 switch, each of which are connected to a respective server farm, so that incoming web requests are handled on a round robin basis (or other SLB technique) before being sent to the cache.

Abstract: The service assignment tool analyzes a service to determine whether the service can execute on a cluster. If the service cannot execute on a cluster, the service is assigned to a single virtual machine. The service assignment tool identifies non-cluster friendly services by performing a static analysis on the bytecode of the service. The bytecode of the service is analyzed by comparing each segment of bytecode to a list of known good and bad coding conventions. If each segment of bytecode in a service meets the good coding convention criteria, then the service is cluster friendly. If one segment of bytecode does not meet the good coding convention criteria, then the entire service is considered to be not cluster friendly.

Abstract: In many messaging systems, message selection filtering involves a significant processing overhead. This includes the parsing and preparing of the message selection filters in readiness for evaluation, as well as the evaluation itself. In systems which involve parsing and prepare processing for expressions that include repeated patterns (such as in many JMS environments), a template-based shortcutting of the conventional processing improves performance. A stored parsing template is used to parse an input sequence such as a received data processing instruction. The parse is used to identify expressions of a first type that can then be implicitly optimized by using a stored evaluation template that is associated with expressions of that type.

Abstract: Embodiments of the present invention provide a method, system and computer program product for computer visualization of drugs for drug interaction information retrieval. In an embodiment of the invention, a method for computer visualization of drugs for drug interaction information retrieval can be provided. The method can include acquiring imagery of multiple different substances and detecting identifying content in each acquired image to determine an identity of each substance corresponding to each acquired image. The method also can include retrieving drug interaction data for each substance using the determined identity and correlating the drug interaction data for at least one of the substances with at least one other of the substances. Finally, the method can include displaying the correlated drug interaction data.

Abstract: Methods, systems, and computer program products for improving the efficiency of data transfer within interconnected components of a virtual network, and in particular components of a single physical computing device, where the components exchange data as if they were communicating over an actual communications network using networking protocols. Data packets to be sent from one component to another are buffered by next-hop address (and optionally by final destination address as well) to improve efficiency of packet delivery.

Abstract: This Sampling Object Cache System (“SOCS”) estimates the size of an in-memory heap-based object cache without the need to serialize every object within the cache. SOCS samples objects at a user-determined rate and then computes a “sample size average” for each type of class—whether a top class, type of top class or non top class. Using these sample size averages, a statistically accurate measure of the overall size of the cache is calculated by adding together the total size of the objects in the cache for each class type.

Abstract: A method, an apparatus, and computer instructions are provided for a context based cache infrastructure to enable subset query over a cached object. Responsive to detecting a query to a root context of a context tree, the tree is traversed for a parent context of a subcontext corresponding to the name and value pair, which is identified by a user in the query. If the parent context caches all query results, the query results are iterated and the remaining name and value pairs are filtered out. However, if the parent context does not cache all query results, the traversing step is repeated for next parent context of the subcontext until a root context is encountered. If a root context is encountered, a query is issued to the database for the name and value pair and the result of the database query is cached in a new context.

Abstract: A method and a system for load balancing media resources among a plurality of servers. A first request can be received from a client requesting a first media processing resource, the request being formatted in accordance with a first protocol. The first request can be morphed into a first morphed request formatted in accordance with a second protocol. A first server that is least loaded can be selected from a first server partition including a plurality of servers, each of which provides the requested first media processing resource. The first morphed request can be forwarded to the first server in accordance with the second protocol.

Abstract: The invention discloses a subscription which is made to receive request messages on one or more topics from a first broker. A subscription is made to receive all messages published at either the first or the second broker to a replyto topic including a bridge identifier. A request message is received from the first broker, the request message specifying a temporary replyto topic on which responses to the request should be published. The temporary reply to topic is modified to specify a bridge identifier. The modified request message is forwarded to a second broker. A response to the request is received from the second broker, the response being published on the replyto topic including the bridge identifier. The bridge receives the response as a result of the subscription to receive all messages published to a topic including the bridge identifier. The response is forwarded to the first broker.

Abstract: A method and system to represent in UML the resources and external data manipulated by a program to be coded in a procedural language is disclosed. The usual UML class diagram is enhanced with two new stereotypes “Program” and “Data Object”. The program stereotypes applied to an operation and its parameters describe the data manipulated, through the logical definition provided by the Data Object. This model will be enhanced in order to introduce classes on which the Resource stereotype is applied. These classes will be connected to the appropriate logical data definition (the Data Object) and used as parameter types of the operation instead of the Data Object. A second set of stereotypes will be applied to the Resource classes to indicate the physical type of the resource (a file, table etc. . . . ) and its characteristics. Parameters that point to a DataObject or a primitive type will be considered as external data.

Abstract: A method, computer program product, and system for receiving a natural language query. The natural language query is parsed into a plurality of base concepts. One or more base concepts chosen from the plurality of base concepts is processed to define a selected report template. The selected report template is selected from a plurality of default report templates. At least one of the base concepts chosen from the plurality of base concepts is mapped to a data field within data via a mapping dictionary. The data includes a plurality of records. A query is executed on at least a portion of the data to identify a result record set. The result record set defines at least a portion of the plurality of records.

Abstract: The Rule Editing Program (REP) displays a rule in a hierarchical tree topology, wherein individual rule portions can be reordered by moving them to different locations within the rule using drag and drop interaction. Dragging and dropping rule portions results in the rule itself being changed accordingly. Additionally, new rule portions can be added to any part of the rule, including the beginning, middle and end. Finally, the entire rule, and individual rule portions, can be expanded all the way out, and collapsed all the way in.

Abstract: A system and computer program product for automatically positioning a computer to an ergonomically correct orientation are presented. An optic sensor at a top of a computer monitor measures a distance and angle to a user's eye. This distance and angle, along with stored anatomical data for a specific user, are used to automatically adjust a base of the computer, thus positioning the computer's keyboard in an ergonomically correct alignment for the specific user.

Abstract: The present invention provides client and server identity validation in an asynchronous request dispatching environment with client-side aggregation. An application server receives an asynchronous include request from a client. A first unique identifier associating the client with the asynchronous include is generated and sent to a results server. A second unique identifier identifying the results server is generated and sent to the application server. Results of the asynchronous include are stored in the results server. The application server sends the first and second unique identifiers to the client, which polls the results server and sends the second unique identifier to the results server. The results server uses the second unique identifier to verify the identity of the client. The results server sends the first unique identifier to the client. The client uses the first unique identifier to validate the identity of the results server.

Abstract: A method of operating a publish/subscribe message broker. A method in accordance with an embodiment includes receiving a message on a topic from a publisher, identifying the message as a start of a sequence of messages, transmitting the message to one or more existing subscribers to the topic, registering a new subscriber to the topic, receiving a further message on the topic from a publisher, and transmitting the further message to the one or more existing subscribers to the topic, but not to the registered new subscriber to the topic. Following the identifying of the message as a start of a sequence of messages, the message broker creates a new sub-topic corresponding to the sequence of messages and subscribes the one or more existing subscribers to the new topic. Any new messages received that relate to the sequence are published to the sub-topic and not the original topic. Any new subscribers to the topic are not subscribed to the sub-topic.

Abstract: Methods, systems, and software for conducting a marketing campaign are described. One or more scores associated with sending offers to proposed contacts are received; and an overall campaign score corresponding to a sum of the scores is optimized to generate optimal assignments of offers to customers. The optimal assignments are included in the marketing campaign.

Abstract: A computer-implemented method of processing recurring events within personal information management (PIM) clients can include executing a plug-in component within a PIM client. The method can include receiving, within the PIM client, an event update including a recurring event identifier that is natively understood by the PIM client and supplemental recurrence data not natively understood by the PIM client. The method further can include, responsive to the plug-in component detecting the supplemental recurrence data and a user input executing the event update, the plug-in component can locate a recurring event corresponding to the recurring event identifier. The method can include the plug-in component updating the recurring event according, at least in part, to the supplemental recurrence data, and storing the recurring event within the PIM client.

Abstract: A system, method and program product for statistically analyzing comments in one or more program code listings. A system is disclosed that includes a process for associating comments to blocks of code at different programmatic levels in a code listing; a process for assigning weights to different code blocks, and a process for generating a statistical analysis at the different programmatic levels of the comments in the code listing.

Abstract: A method, apparatus, and computer instructions are provided for preserving relationships of mapped views in an object model. A database model, including database tables and views is examined and a foreign key constraint list is derived for the view. A set of objects for an object model is generated from the database model. Based on the derived foreign key list, relationships are generated between objects mapped to the view and objects mapped to the target table of the view's derived foreign key.

Abstract: The invention comprises a financial management program and one or more databases containing information on merchants and their products. The program interacts with the databases containing the product information to return the prices and availability of goods located in a designated area. The program compares prices from various merchants and displays the results to the consumer. The program also allows the consumer to compare the prices with a household budget an financial data, while keeping track of trends in the consumer's spending and buying habits.