Abstract: A computer-implemented method, a computerized system and a product for providing a cluster of replicated servers. The method performed by a computerized server in a cluster of servers, wherein the cluster of servers are executing replicated instances of an application, wherein the replicated instances are configured to perform the same processing of the same input, comprising: detecting a message loss in the server; electively determining a responsive action to the message loss; and notifying the cluster of servers of the responsive action determined by the server, whereby other servers of the cluster of servers are able to mimic operation of the server by simulating the responsive action.

Abstract: Systems and methods for optimal data storage are provided. The method comprises storing data objects that are most closely related in a storage container by modeling a constraint satisfaction problem for placement of said data objects in one or more storage containers, wherein a weight is assigned to an edge connecting two data objects based on an association defining relationships between the two data objects connected by said edge, taking into account certain penalties for placing multiple copies of the same object in the one or more storage containers, and wherein a storage container comprises a logical or physical storage area as a unit of storage.

Abstract: A method for server replication in a network environment is provided. The primary server provides a first service to a client. If the first service involves interaction with a non-deterministic data source, the primary server performs the interaction and provides information about the interaction to a secondary server that is a replica of the primary server. The secondary server uses the information about the interaction to synchronize the secondary server with the primary server.

Abstract: A method of routing dependent messages sent from a source node. The method comprises routing a plurality of messages including a plurality of dependent messages from a source node for processing by a group of a plurality of processing nodes, optionally while managing a failure recovery mechanism and complying with message dependencies. Each message having a weight, each dependent message is routed while at least one dependency thereof is complied with, acquiring a plurality of acknowledge notifications to at least some of the plurality of messages from the plurality of processing nodes, calculating, at the source node using a processor, a message load of each of the plurality of processing nodes according to the weight of respective messages of plurality of messages which are sent thereto and respective acknowledge notifications of the plurality of acknowledge notifications which are sent therefrom. The routing is performed according to the respective message load.

Abstract: A computer-implemented method, a computerized system and a product for providing a cluster of replicated servers. The method performed by a computerized server in a cluster of servers, wherein the cluster of servers are executing replicated instances of an application, wherein the replicated instances are configured to perform the same processing of the same input, comprising: detecting a message loss in the server; electively determining a responsive action to the message loss; and notifying the cluster of servers of the responsive action determined by the server, whereby other servers of the cluster of servers are able to mimic operation of the server by simulating the responsive action.

Abstract: A group communication service maintains synchronization limiting the group to the processing entities that receive a plurality of source message streams from sources outside the group and transmit delivery message streams to destinations that are outside the group. Synchronization of message ordering and other state synchronization directives are communicated from a leader exclusively to other group members. Once having declared themselves to be synchronized and active by a reply protocol message, which is an opaque object and is interpreted by an application layer, the processing entities independently process the source message streams in accordance with the order specified in the directives, without requiring additional sequencing information from another processing entity.

Abstract: Systems and methods for optimal data storage are provided. The method comprises storing data objects that are most closely related in a storage container by modeling a constraint satisfaction problem for placement of said data objects in one or more storage containers, wherein a weight is assigned to an edge connecting two data objects based on an association defining relationships between the two data objects connected by said edge, taking into account certain penalties for placing multiple copies of the same object in the one or more storage containers, and wherein a storage container comprises a logical or physical storage area as a unit of storage.

Abstract: A method, system and computer program product for transmitting data entities, the method includes: receiving data entities to be transmitted over multiple channels; and preventing a transmission of data entities that are to be transmitted over congested channels while transmitting data entities that are to be are to be transmitted over non-congested channels.

Abstract: A group communication service is adapted for multi-point to multi-point communication. Identical message streams are received by a tier of processing servers from a tier of sources. The processing servers deliver the message streams to a tier of destinations. The group communication service enables the processing servers to deal with newly added message streams and deleted message streams while preserving total ordering.

Abstract: A method for communication in a computing environment is provided. The method comprises transmitting a first data packet to one or more receiving units in a first set; receiving one or more acknowledgement messages (ACKs) from one or more receiving units in a second set, wherein the second set is a subset of the first set; removing the first data packet from a queue, in response to receiving at least one ACK from N receiving units within a time threshold; and re-transmitting the first data packet to the first set, in response to failing to receive at least one ACK from N receiving units within the time threshold.

Abstract: A publish/subscribe system includes a publisher and multiple subscribers. The published includes a packet generator to generate a packet with at least one of: a fixed length flow label labeling a message or a message bundle and a fixed length node label labeling at least one message bundle and indicating at least one of the nodes of a topic tree to which the flow label belongs. Each received includes a user selection data holder and a message selector. The data holder stores a selection of topics of interest to a user from the topic tree. The message selector receives a packet having a header and data, determines which flow and/or node labels are incorporated in the header and accepts the data if the user selection data holder indicates that the labels within the header are of interest to the user.

Abstract: Methods, systems, and products are disclosed for filtering application messages in a high speed, low latency data communications environment that include: receiving an application message in a transport engine of a message receiving device from a message transmitting device, the application message having a message contents label and a collision indicator, the message contents label representing contents of the application message, the collision indicator specifying whether the message contents label represents different contents in at least one other application message; determining, by the transport engine, whether contents of the application message satisfy a transport layer constraint in dependence upon the message contents label; providing, by the transport engine, the application message and the collision indicator to messaging middleware of the message receiving device if the contents of the application message satisfy the transport layer constraint; and administering the application message, by the me

Abstract: A method for server replication in a network environment is provided. The primary server provides a first service to a client. If the first service involves interaction with a non-deterministic data source, the primary server performs the interaction and provides information about the interaction to a secondary server that is a replica of the primary server. The secondary server uses the information about the interaction to synchronize the secondary server with the primary server.

Abstract: Efficient Light-Weight Multicasting Communication Protocol A method for communication in a computing environment is provided. The method comprises transmitting a first data packet to one or more receiving units in a first set; receiving one or more acknowledgement messages (ACKs) from one or more receiving units in a second set, wherein the second set is a subset of the first set; removing the first data packet from a queue, in response to receiving at least one ACK from N receiving units within a time threshold; and re-transmitting the first data packet to the first set, in response to failing to receive at least one ACK from N receiving units within the time threshold.

Abstract: A method, system and computer program product for transmitting data entities, the method includes: receiving data entities to be transmitted over multiple channels; and preventing a transmission of data entities that are to be transmitted over congested channels while transmitting data entities that are to be are to be transmitted over non-congested channels

Abstract: A group communication service is adapted for multi-point to multi-point communication. Identical message streams are received by a tier of processing servers from a tier of sources. The processing servers deliver the message streams to a tier of destinations. The group communication service enables the processing servers to deal with newly added message streams and deleted message streams while preserving total ordering.

Abstract: A group communication service maintains synchronization limiting the group to the processing entities that receive a plurality of source message streams from sources outside the group and transmit delivery message streams to destinations that are outside the group. Synchronization of message ordering and other state synchronization directives are communicated from a leader exclusively to other group members. Once having declared themselves to be synchronized and active by a reply protocol message, which is an opaque object and is interpreted by an application layer, the processing entities independently process the source message streams in accordance with the order specified in the directives, without requiring additional sequencing information from another processing entity.

Abstract: Methods, systems, and products are disclosed for filtering application messages in a high speed, low latency data communications environment that include: receiving an application message in a transport engine of a message receiving device from a message transmitting device, the application message having a message contents label and a collision indicator, the message contents label representing contents of the application message, the collision indicator specifying whether the message contents label represents different contents in at least one other application message; determining, by the transport engine, whether contents of the application message satisfy a transport layer constraint in dependence upon the message contents label; providing, by the transport engine, the application message and the collision indicator to messaging middleware of the message receiving device if the contents of the application message satisfy the transport layer constraint; and administering the application message, by the me

Abstract: Methods, apparatus, and products are disclosed for reliable messaging using message streams in a high speed, low latency data communications environment that include: receiving, in a transport engine of a message receiving device from an active message sending device, active transport packets on an active message stream established from the active message sending device to the message receiving device, the active message sending device encapsulating active application messages in the active transport packets; identifying, by the transport engine, a missing active application message from the active message sending device; and requesting, by the transport engine from a backup message sending device, transmission of a backup application message that corresponds to the missing active application message, the backup message sending device encapsulating backup application messages in backup transport packets for transmission on a backup message stream, each backup application message representing a duplicate of a

Abstract: A method for migrating data transmitted from a transmitter to a receiver over a first stream to a second stream in a reliable multicast system is provided. The method comprises transmitting a first message from the transmitter to the receiver over the first stream to notify the receiver that a first data flow transmitted on the first stream will be transmitted on the second stream. The transmitter transmits a second message to the receiver over the second stream after a second threshold has expired. The receiver tunes to the second stream based on the second message. The transmitter transmits a third message to the receiver over the first stream after a third threshold has expired to notify the receiver that transmission of the first data flow over the first stream will be terminated. The transmitter then transmits a fourth message from the transmitter to the receiver over the second stream after a fourth threshold has expired.