Method for operating an arrangement of a plurality of computers in the event of a computer failure

• US 7,574,620 B2
• Filed: 08/02/2006
• Issued: 08/11/2009
• Est. Priority Date: 02/02/2004
• Status: Expired due to Fees

• Current Assignee: Fujitsu Siemens Computers GmbH (Fujitsu Limited)
• Original Assignee: Fujitsu Siemens Computers GmbH (Fujitsu Limited)
• Inventors: Hartung, Klaus
• Primary Examiner(s): Beausoliel; Robert
• Assistant Examiner(s): Lottich; Joshua P
• Application Number:
  US11/497,525
• Publication Number:
  US 20070038885A1
• Time in Patent Office: 1,105 Days
• Field of Search: 714/4, 714/38, 714/43, 714/47, 709/221, 709/223, 709/225
• US Class Current: 714/4.21
• CPC Class Codes: G06F 11/143 with loss of software functionality
• CPC Class Codes: G06F 11/1438 Restarting or rejuvenating
• CPC Class Codes: G06F 11/2025 using centralised failover control functionality
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• 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 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1029 using data related to the state of servers 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 established for operating an arrangement of a plurality of computers that are configured to execute software units. Continuous monitoring of the plurality of computers and intervening with any failed computers is performed via a monitoring computer. The software units are assigned a weighting in accordance with their importance, and the method steps are invoked in the event of a computer failing. At the monitoring time, information regarding the failure of all of the failed computers and software units affected is thereby collected. The software units, that are affected by the failure or failures, are transferred to other computers. The transfer is determined via the weighting of the software units.

Claims

1. A method for operating a server farm including an arrangement of a plurality of computers that are configured to execute a plurality of software units, and an additional monitoring computer to conduct repeated monitoring of the status of the plurality of computers and the intervening of the plurality of computers, wherein the software units are assigned a weighting according to a selected importance criteria, the method comprising:

when at least one computer has been determined as failing, collecting information at scheduled monitoring times and in relation to which of the computers have failed and which software units are thereby affected;

determining a plurality of possible transfer scenarios for transferring the affected software units from failed computers to active computers including terminating active software units based on the weighting so as to establish free capacity for transferring software units from computers that have failed, wherein the weighting is achieved via criteria comprising a plurality of attributes including;

a first attribute that specifies a measure of the power to have other active software units terminated thereby creating free capacity, and a second attribute that specifies a measure of the resistance to the request for termination by other software units that are to be transferred from a failed computer and each transfer scenario indicates which software units are to be transferred to which active computers;

assessing the possible transfer scenarios based upon the weighting of the software units;

selecting a transfer scenario based on the assessment of possible transfer scenarios; and

executing the selected transfer scenario.

2. The method of claim 1, further comprising:

transferring the affected software units to a reserve computer with sufficient free capacity.

3. The method of claim 1, wherein the attributes of the software units that are active in the computer arrangement and of the software units that are affected by failed computers are considered in selecting the transfer scenario such that the selected transfer scenario is optimal to the software units in the overall arrangement of the plurality of computers.

4. The method of claim 3, wherein the possible transfer scenarios are sorted according to predetermined criteria prior to selection of the optimal transfer scenario.

5. The method of claim 3, further comprises using a greedy algorithm to select the optimal transfer scenario from the possible transfer scenarios.

6. The method of claim 1, wherein each of the software units includes an operating system and applications.

7. A system comprising:

a plurality of computers that execute a plurality of software units, and a monitoring computer to repeatedly monitor the status of the plurality of computers and to intervene with the plurality of computers, in response to at least one of the plurality of computers failing, wherein the monitoring computer is configured to execute the following steps;

collect information at scheduled monitoring times and in relation to which of the computers have failed and which software units are thereby affected;

determine a plurality of possible transfer scenarios for transferring the affected software units from failed computers to active computers including terminating active software units based on the weighting so as to establish free capacity for transferring software units from computers that have failed, wherein the weighting is achieved via criteria comprising a plurality of attributes including;

a first attribute that specifies a measure of the power to have other active software units terminated thereby creating free capacity, and a second attribute that specifies a measure of the resistance to the request for termination by other software units that are to be transferred from a failed computer and each transfer scenario indicates which software units are to be transferred to which active computers;

assess the possible transfer scenarios based upon the weighting of the software units;

select a transfer scenario based on the assessment of possible transfer scenarios; and

execute the selected transfer scenario.