Dynamically configurable fault tolerance in autonomic computing with multiple service points

• US 7,111,188 B2
• Filed: 07/31/2003
• Issued: 09/19/2006
• Est. Priority Date: 07/31/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Mukherjee, Maharaj
• Primary Examiner(s): Bonzo, Bryce P.
• Assistant Examiner(s): MASKULINSKI, MICHAEL C
• Application Number:
  US10/604,585
• Publication Number:
  US 20050027829A1
• Time in Patent Office: 1,146 Days
• Field of Search: 714/1, 370254-256
• US Class Current: 714/1
• CPC Class Codes: G06F 2209/505 Clust
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1021 based on client or server locations
• CPC Class Codes: H04L 67/1023 based on a hash applied to IP addresses or costs
• CPC Class Codes: H04L 67/1034 Reaction to server failures by a load balancer
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A method is described for configuring a system having a plurality of processors to provide the system with at least one cluster of processors, where each cluster has one service point. A distance is computed from each processor to other processors in the system. A plurality of total distances is then computed, where each total distance is associated with one processor. A minimum total distance is determined from the plurality of total distances. One processor is assigned to be the service point; this processor is the processor having the minimum total distance associated therewith.

Claims

1. A method for configuring a system having a plurality of processors to provide the system with at least one cluster of processors, each cluster having one service point, the method comprising the steps of:

computing a distance from each processor to other processors in the system;

computing a plurality of total distances, where each total distance is associated with one processor;

determining a minimum total distance from the plurality of total distances;

assigning as the service point the processor having the minimum total distance associated therewith;

partitioning the system into a plurality of clusters, the partitioning comprising;

sorting the processors in accordance with the total distance associated with each processor;

assigning each processor to one of two clusters;

determining a minimum total distance for the processors in each cluster in accordance with the plurality of total distances associated with the processors in said cluster; and

assigning as the service point for each cluster the processor having the minimum total distance associated therewith in said cluster,

subdividing one of said two clusters into two subdivided clusters, thereby partitioning the system into three clusters;

determining a minimum total distance for the processors in each of said three clusters in accordance with the plurality of total distances associated with the processors in said three clusters;

assigning the processors to said three clusters in accordance with the minimum total distance; and

assigning as the service point for each of said three clusters the processor having the minimum total distance associated therewith in said cluster.

2. A method according to claim 1, wherein said configuring is performed dynamically when a processor is added to the system.

3. A method according to claim 1, wherein said configuring is performed dynamically when a processor is removed from the system.

4. A method according to claim 3, wherein the partitioning of the system is dynamically changed when a processor is removed from the system.

5. A method according to claim 1, further comprising the step of assigning another processor as a backup service point.