Effective selection of a virtual machine to be moved outside influence range of a failure

Patent metadata

Field	Value
Publication number	US-11354201-B2
Application number	US-201916435606-A
Country	US
Kind code	B2
Filing date	Jun 10, 2019
Priority date	Jun 14, 2018
Publication date	Jun 7, 2022
Grant date	Jun 7, 2022

Abstract

A method performed by an apparatus is provided. The apparatus extracts a virtual machine before starting operation from among virtual machines within a range of influence of a failure upon detection of the failure. With reference to a storage unit storing history information concerning operation of a virtual machine, the apparatus generates first information corresponding to time intervals of operation start time of the extracted virtual machine, second information corresponding to movement required time required to move the extracted virtual machine out of the range of influence of the failure, and third information corresponding to recovery required time required for recovery of the failure detected to have occurred. The apparatus determines whether to move the extracted virtual machine out of the range of influence of the failure, based on the generated first, second, and third information.

First claim

What is claimed is:

1. A non-transitory, computer-readable recording medium having stored therein a program for causing a computer to execute a process comprising:
extracting a virtual machine before starting operation from among virtual machines within a range of influence of a failure upon detection of the failure;
with reference to a memory for storing history information concerning operation of the virtual machine, generating first information corresponding to time intervals of operation start time of the extracted virtual machine, second information corresponding to movement required time required to move the extracted virtual machine out of the range of influence of the failure, and third information corresponding to recovery required time required for recovery of the failure detected to have occurred; and
determining whether to move the extracted virtual machine out of the range of influence of the failure, based on the generated first, second, and third information.

2. The non-transitory, computer-readable recording medium of claim 1 , wherein the generating includes:
referring to the memory for storing the history information to specify one or more time intervals in the past for the extracted virtual machine,
generating, as the first information, information indicating a percentage for each of the specified time intervals to whole the specified time intervals,
referring to the memory for storing the history information to specify movement required times in the past for the extracted virtual machine,
generating, as the second information, information indicating the percentage for each of the specified movement required times to whole the specified movement required times,
referring to the memory for storing the history information to calculate a time difference between the recovery required time upon a past occurrence of the failure and a predicted time for the recovery required time predicted upon the past occurrence of the failure, for each of failures occurred in the past, and
generating, as the third information, information indicating a percentage for each of the calculated time differences to whole the calculated time differences.

3. The non-transitory, computer-readable recording medium of claim 2 , wherein the determining includes:
calculating an elapsed time from a previous operation start time to a current time for the extracted virtual machine,
calculating, from the first and third information, a first probability at which a time obtained by subtracting the calculated elapsed time from a selected one of the time intervals is shorter than a time obtained by adding the calculated time difference to the predicted time for the recovery required time predicted for the failure detected to have occurred,
calculating, from the first and second information, when the calculated first probability is not less than a first threshold, a second probability at which the time obtained by subtracting the calculated elapsed time from the selected one of the time intervals is shorter than the movement required time, and
moving, when the calculated second probability is not more than a second threshold, the extracted virtual machine out of the range of influence of the failure.

4. The non-transitory, computer-readable recording medium of claim 2 , wherein
the generating includes generating the first information, the second information, and the third information for each of the virtual machines when there is more than one virtual machine extracted; and
the determining includes:
calculating, when there is more than one virtual machine determined to be moved out of the range of influence of the failure, priority information for each of the virtual machines from the first probability, and
determining to move the virtual machines out of the range of influence of the failure sequentially from a virtual machine having a larger value of the priority information calculated.

5. The non-transitory, computer-readable recording medium of claim 4 , the process further comprising when a plurality of virtual machines before operation have been extracted, performing again the determining of whether to move the extracted virtual machines out of the range of influence of the failure, upon completion of the movement of each of the plurality of virtual machines out of the range of influence of the failure.

6. The non-transitory, computer-readable recording medium of claim 4 , the process further comprising:
when a plurality of virtual machines before operation have been extracted, calculating the number of the virtual machines that may be moved out of the range of influence of the failure when moving more than one virtual machine in a predetermined order by repeating the determining of whether to move the extracted virtual machines out of the range of influence of the failure, while assuming that the plurality of virtual machines are moved one by one in the predetermined order; and
determining to move more than one virtual machine in the predetermined order when the calculated number is not less than a threshold.

7. The non-transitory, computer-readable recording medium of claim 1 , wherein:
the generating includes referring to the memory for storing the history information to generate fourth information indicating addition required time required to add resources of a physical machine in which the extracted virtual machine is operated; and
the determining includes determining, based on the generated first and fourth information, whether to add the resources of the physical machine in which the extracted virtual machine is operated, when determining that the extracted virtual machine is not to be moved out of the range of influence of the failure.

8. The non-transitory, computer-readable recording medium of claim 1 , wherein
the generating the first information, the second information, and the third information includes:
specifying a physical machine in which the extracted virtual machine is operated,
specifying the number of virtual machines operated in the specified physical machine, and
referring to the memory for storing history information to specify history information corresponding to the specified number, among the history information, to generate the first information, the second information, and the third information.

9. A failure handling method performed by a processor included in a failure handling apparatus, the failure handling method comprising:
extracting a virtual machine before starting operation from among virtual machines within a range of influence of a failure upon detection of the failure;
referring to a memory for storing history information concerning operation of a virtual machine to generate first information corresponding to time intervals of operation start time of the extracted virtual machine, second information corresponding to movement required time required to move the extracted virtual machine out of the range of influence of the failure, and third information corresponding to recovery required time required for recovery of the failure detected to have occurred; and
determining whether to move the extracted virtual machine out of the range of influence of the failure, based on the generated first, second, and third information.

10. The failure handling method of claim 9 , wherein the generating includes:
referring to the memory for storing the history information to specify a time interval in the past for the extracted virtual machine;
generating, as the first information, information indicating a percentage for each of the specified time intervals to whole the specified time intervals;
referring to the memory for storing the history information to specify movement requi

Assignees

• Fujitsu Ltd

Inventors

• Kondo Reiko
• Watanabe Yukihiro
• Asaoka Masahiro
• Uchiumi Tetsuya
• Iikura Fumi
• Suzuki Kazuhiro
• Okuno Shingo
• Saito Yuji

Classifications

• G06F2009/45591

  Monitoring or debugging support · CPC title

• G06F11/142 Primary

  Reconfiguring to eliminate the error (group management mechanisms in a peer-to-peer network H04L67/1044) · CPC title

• G06F9/45558

  Hypervisor-specific management and integration aspects · CPC title

• G06F2009/45562

  Creating, deleting, cloning virtual machine instances · CPC title

• G06F2009/45575

  Starting, stopping, suspending or resuming virtual machine instances · CPC title