Method and apparatus for load balancing a distributed processing system

• US 20050125798A1
• Filed: 09/07/2004
• Published: 06/09/2005
• Est. Priority Date: 04/24/2001
• Status: Abandoned Application

• Current Assignee: Diane Lee Peterson
• Original Assignee: Diane Lee Peterson
• Inventors: Peterson, Diane L.
• Application Number:
  US10/937,751
• Publication Number:
  US 20050125798A1
• Time in Patent Office: Days
• US Class Current: 718/105
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system
• CPC Class Codes: H04L 45/00 Routing or path finding of packets in data switching networks routing or path finding in wireless networks H04W40/00
• CPC Class Codes: H04L 45/70 Routing based on monitoring results
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/29 using a combination of thresholds
• 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 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/10015 Access to distributed or replicated servers, e.g. using brokers
• CPC Class Codes: H04L 69/08 Protocols for interworking; Protocol conversion
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

Method and apparatus for load balancing a distributed processing system. A method for load balancing a distributed processing system having a plurality of processing nodes associated therewith is provided. A portion of a process is first received at one of the processing nodes for processing of that portion thereat. The amount of processing at that node is then determined that will be required to complete the received portion of the process as a processing load on that one processing node. A determination is then made if the amount of processing exceeds a predetermined threshold. If the amount of processing exceeds the predetermined threshold, then information is transferred to other nodes in the network that the one of the processing nodes is unavailable for further processing.

Claims

1. :

A method for load balancing a distributed processing system having a plurality of processing nodes associated therewith, comprising the steps of;

receiving a portion of a process at one of the processing nodes for processing of that portion thereat;

determining the amount of processing that will be required to complete the received portion of the process as a processing load on that one processing node based on the transaction being processed;

wherein the process is a portion of a larger process that is comprised of a plurality of discrete processes, wherein the one of the nodes is operable to operate on a plurality of processes and the step of determining is operable to determine the amount of time that is required for the entire process to be completed;

determining if the amount of processing exceeds a predetermined threshold; and

if the amount of processing exceeds the predetermined threshold, then transferring information to other nodes in the network that the one of the processing nodes is unavailable for further processing.

2. :

The method of claim 1, wherein the processing at the one of the nodes comprises a process that requires a stream of data to be received from another one of the nodes in the system.

3. :

The method of claim 2, wherein the another one of the nodes is the one originating the process that is sent to the one of the nodes.

4. :

The method of claim 1, wherein the step of determining if the amount of processing exceeds the predetermined threshold is facilitated at another load monitoring node on the network remote from the one node and the step of determining the amount of processing is operable to push this information to the load monitoring node and the load monitoring node is operable to transfer information to other nodes in the network that the one of the processing nodes is unavailable for further processing.

5. :

The method of claim 4, wherein the one processing node is operable notify the load monitoring node of any change in the processing load thereon and, when the step of determining if the processing load is above the predetermined threshold determines that it is below the predetermined threshold, then transferring information to other nodes in the network that the one of the processing nodes is available for further processing.

6. :

The method of claim 4, wherein the load monitor node is operable to compare the amount of time that the one processing node is at a processing load that is above the predetermined threshold level to a predetermined expected level and, if it exceeds this time, taking a predetermined action.

7. :

The method of claim 6 wherein the predetermined action is to send a request for status to the one processing node.

8. :

A method for load balancing a distributed processing system having a plurality of processing nodes associated therewith, comprising the steps of;

originating a transaction at an originating one of the processing nodes, which transaction is comprised of a plurality of discrete processes that are to be carried out a plurality of the processing nodes on the network and in a predetermined order along a transaction path, and associating a predetermined transaction ID with the transaction that defines the transaction and the associated path uniquely on the network;

transferring a message from each of the processing nodes in the transaction path to the next of the processing nodes in the transaction path after processing the portion of the transaction associated therewith;

at least one of the processing nodes, in response to receiving the transferred message, determining the amount of processing that will be required to complete the portion of the transaction associated therewith as a processing load on that one processing node;

determining if the amount of processing exceeds a predetermined threshold; and

if the amount of processing exceeds the predetermined threshold, then transferring information to at least one other of the nodes in the network other than the at least one of the processing nodes that the at least one of the processing nodes is unavailable for further processing.

9. :

The method of claim 8, wherein the processing at the one of the nodes comprises a process that requires a stream of data to be received from another one of the nodes in the system.

10. :

The method of claim 9, wherein the at least one other of the nodes is the originating one of the processing nodes.

11. :

The method of claim 8, wherein the step of determining is based upon the transaction being processed.

12. :

The method of claim 11, wherein the at least one of the nodes is operable to operate on a plurality of processes that comprise the transaction and the step of determining is operable to determine the amount of time that is required for the entire process to be completed.

13. :

The method of claim 8, wherein the step of determining if the amount of processing exceeds the predetermined threshold is facilitated at another load monitoring node on the network remote from the at least one node and the step of determining the amount of processing is operable to push this information to the load monitoring node and the load monitoring node is operable to transfer information to other nodes in the network that the at least one of the processing nodes is unavailable for further processing.

14. :

The method of claim 13, wherein the at least one processing node is operable notify the load monitoring node of any change in the processing load thereon and, when the step of determining if the processing load is above the predetermined threshold determines that it is below the predetermined threshold, then transferring information to other nodes in the network that the one of the processing nodes is available for further processing.

15. :

The method of claim 13, wherein the load monitor node is operable to compare the amount of time that the at least one processing node is at a processing load that is above the predetermined threshold level to a predetermined expected level and, if it exceeds this time, taking a predetermined action.

16. :

A method for parameterizing a distributed processing system having a plurality of processing nodes associated therewith, comprising the steps of;

receiving a portion of a process at one of the processing nodes for processing of that portion thereat in accordance with a set of associated parameters defining the portion;

monitoring the operation of the overall process for a given transaction involving a process across the distributed processing system and determining if an alteration of the process is required in accordance with predetermined criteria; and

if an alteration of the process is required, then transferring new parameters to the processing node.