Systems and methods for synchronizing data between communication devices in a networked environment

• US 8,645,541 B1
• Filed: 03/16/2010
• Issued: 02/04/2014
• Est. Priority Date: 12/12/2003
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: Open Invention Network LLC
• Inventors: Jones, Dan, Vernon, Todd, Kriese, Jeffrey
• Primary Examiner(s): Serrao, Ranodhi
• Application Number:
  US12/724,995
• Time in Patent Office: 1,421 Days
• Field of Search: 709/226, 709/228, 711/100, 718/101
• US Class Current: 709/226
• CPC Class Codes: H04L 41/0213 Standardised network management protocols, e.g. simple network management protocol [SNMP]
• CPC Class Codes: H04L 49/90 Buffering arrangements
• CPC Class Codes: H04L 65/1101 Session protocols
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/1004 Server selection for load balancing
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 67/61 taking into account QoS or priority requirements
• CPC Class Codes: H04L 69/28 Timers or timing mechanisms used in protocols

Abstract

A network communication system and a source device thereof. The source device may include a cluster manager that groups a plurality of connections between the source device and the plurality of destination devices into clusters, such as performance clusters. The cluster manager may also remove one or more of the destination devices that do not respond within a timeout threshold from their clusters, and when the removed one or more destination devices respond to the timeout threshold, the cluster manager may provide one or more of the plurality of destination devices full-buffer updates until such point as a recent average latency performance of a responding destination device is at a level that merits insertion back into one of the clusters.

Claims

1. A source device operable to send data to a plurality of destination devices, the source device comprising:

a processor and a memory unit; and

a cluster manager, the cluster manager structured to;

group a plurality of connections between the source device and the plurality of destination devices into clusters; and

remove one or more of the destination devices that do not respond within a timeout threshold from their clusters, and when the removed one or more destination devices respond to the timeout threshold, provide the removed one or more of the plurality of destination devices full-buffer updates until a recent average latency performance of the removed one or more destination devices is at a level that merits insertion back into one of the clusters.

2. The source device of claim 1, wherein the cluster manager groups the plurality of connections into performance clusters based on at least one of performance similarities, client service priorities, and resource priorities.

3. The source device of claim 1, further comprising a plurality of connection interfaces and a plurality of communication connections, wherein the source device is interfaced with the plurality of destination devices via the plurality of connection interfaces and the plurality of communication connections.

4. The source device of claim 3, further comprising a source data buffer configured to store data, and wherein the source device is structured to share the data stored in the source data buffer with the plurality of destination devices interfaced with the source device via the plurality of connection interfaces.

5. The source device of claim 3, further comprising a plurality of synchronization mechanisms coupled with the plurality of connection interfaces, wherein the cluster manager is configured to assign a synchronization mechanism to each of the clusters.

6. The source device of claim 5, wherein each of the plurality of synchronization mechanisms is configured to provide computations and protocols needed to communicate data over the plurality of connections.

7. The source device of claim 1, wherein the cluster manager is configured to at least one of create fewer clusters when it is determined that resource priorities are a priority and create more clusters when it is determined that client service is a priority.

8. The source device of claim 1, wherein the cluster manager is further configured to detect a change in performance capabilities for one of the plurality of connections and to assign a connection for the one of the plurality of connections to another cluster based on the change in performance capabilities.

9. The source device of claim 1, wherein the cluster manager is further configured to detect a new connection, determine a performance of the new connection, and add the new connection to a cluster based on the performance of the new connection.

10. A network communication system for sending data from an original source device to a plurality of destination devices comprising:

a processor and a memory unit;

an intermediate source device, wherein the intermediate source device comprises;

a cluster manager configured to determine subsets of connections from a set of connections and group each of the subsets in a distinct cluster, and

wherein one or more of the destination devices that do not respond within a timeout threshold are removed from their distinct clusters, but when the removed one or more destination devices respond to the timeout threshold, the removed one or more destination devices receive full-buffer updates until a recent average latency performance of the removed one or more destination devices is at a level that merits insertion back into one of the distinct clusters.

11. The network communication system of claim 10, wherein each connection in each subset has similar performance capabilities with the other connections in the same subset.

12. The network communication system of claim 11, wherein the performance capabilities are based on at least one of performance similarities, client service priorities, and resource priorities.

13. The network communication system of claim 12, wherein the performance similarities for each subset of connections is determined by a similarity in connection bandwidth.

14. The network communication system of claim 12, wherein the performance similarities for each subset of connections is determined by a similarity in connection security.

15. The network communication system of claim 12, wherein the performance similarities for each subset of connections is determined by a similarity in connection error rate.

16. The network communication system of claim 10, further comprising a plurality of synchronization mechanisms, wherein the cluster manager is configured to assign a synchronization mechanism to each of the clusters.

17. The network communication system of claim 16, wherein each of the plurality of synchronization mechanisms is configured to provide computations and protocols needed to communicate data over the set of connections.

18. The network communication system of claim 10, further comprising an intermediate source data buffer configured to store data, and wherein the intermediate source device is configured to share the data stored in the source data buffer with the plurality of destination devices.

19. The network communication system of claim 10, wherein the cluster manager is further configured to detect a change in performance capabilities for a connection in one cluster and to assign the connection to another cluster based on the change in performance capabilities.

20. A source device structured to send data to a plurality of destination devices comprising:

a processor and a memory unit;

a cluster manager configured to;

determine subsets of connections from a set of connections, and group each of the subsets in a distinct cluster, and

wherein one or more of the destination devices that do not respond within a timeout threshold are removed from their distinct clusters, but when the removed one or more destination devices respond to the timeout threshold, the removed one or more destination devices receive full-buffer updates until a recent average latency performance of the one or more destination devices is at a level that merits insertion back into one of the distinct clusters.