Operation management device and method

Patent metadata

Field	Value
Publication number	US-10177971-B2
Application number	US-201514625833-A
Country	US
Kind code	B2
Filing date	Feb 19, 2015
Priority date	Aug 30, 2012
Publication date	Jan 8, 2019
Grant date	Jan 8, 2019

Abstract

An operation management device includes a processor that executes a procedure including: for plural nodes that includes plural nodes operating as virtual computers on one or more computers and that plural nodes that have a dependency relationship, deriving group information indicating the plural nodes having the dependency relationship on a basis of communication information from when the inter-node communication; and based on location information of the respective plural nodes included in the derived group information, determining operation information indicating that plural nodes included in the group information operate on a single computer, or operation information indicating that the plural nodes included in the group information respectively operate on plural computers, for the plural nodes included in the group information.

First claim

What is claimed is:

1. An operation management device, comprising:
a processor that executes a process, the process comprising:
for a plurality of nodes that includes a plurality of nodes operating as virtual computers on one or more physical computers and that a plurality of nodes that have a dependency relationship and that perform mutual inter-node communication during operation, calculating a degree of similarity between the plurality of nodes on a basis of communication information from when the inter-node communication is performed and deriving a virtual group that includes a subset of the plurality of nodes between whom the calculated degree of similarly exceeds a specific threshold, the communication information including physical location information of the plurality of nodes, which indicates one or more physical computers on which the respective ones of the plurality of nodes operate;
based on the physical location information for each of the plurality of nodes included in the derived virtual group, determining operation information indicating that a plurality of nodes included in the virtual group operate on a single physical computer, or operation information indicating that the plurality of nodes included in the virtual group respectively operate on a plurality of physical computers; and
based on the determined operation information, if a first physical computer, on which a first node included in the virtual group, operate, is different from a second physical computer, on which a second node included in the virtual group and having a degree of similarity to the first node that exceeds the predetermined threshold, operate, moving the first node from the first physical computer to the second physical computer.

2. The operation management device of claim 1 , wherein, the communication information between a plurality nodes includes a communication event count of mutual communication performed between the plurality of nodes.

3. The operation management device of claim 1 , wherein, the operation information indicates whether or not a subset of nodes included in the virtual group operates on a different physical computer from nodes included in the virtual group other than the subset of nodes.

4. The operation management device of claim 1 , the process further comprising:
adding a node to the virtual group that includes nodes whose degree of similarity to the added node exceeds the specific threshold.

5. The operation management device of claim 4 , wherein, as the degree of similarity, a Euclidean distance is found between the added node and one node out of the plurality of nodes.

6. The operation management device of claim 1 , further comprising a storage section that stores a virtual group information and the operation information as a table, the virtual group information indicating which subset of the plurality of nodes form a common virtual group.

7. The operation management device of claim 4 , the process further comprising:
based on the virtual group of the added node and the operation information of the added node, when a physical computer on which the added note operates does not match a physical computer on which a node exhibiting a degree of similarity of the specific value or greater operates, moving the added node to the physical computer on which the node exhibiting a degree of similarity of the specific value or greater operates.

8. The operation management device of claim 1 , the process further comprising:
when a physical computer including at least one node is in a critical state that is problematic for operation, moving a node that operates on the critical state computer to a physical computer that is different from the critical state computer and that is associated with the operation information of the virtual group including the node.

9. An operation management method, comprising:
for a plurality of nodes that includes a plurality of nodes operating as virtual computers on one or more physical computers and that includes a plurality of nodes that have a dependency relationship and that perform mutual inter-node communication during operation, calculating a degree of similarity between the plurality of nodes on a basis of communication information from when the inter-node communication is performed and deriving a virtual group that includes a subset of the plurality of nodes between whom the calculated degree of similarly exceeds a specific threshold, the communication information including physical location information of the plurality of nodes, which indicates one or more physical computers on which the respective ones of the plurality of nodes operate;
by a processor, based on the physical location information for each of the plurality of nodes included in the derived virtual group, determining operation information indicating that a plurality of nodes included in the virtual group operate on a single physical computer, or operation information indicating that the plurality of nodes included in the virtual group respectively operate on a plurality of computers; and
based on the determined operation information, if a first physical computer, on which a first node included in the virtual group, operate, is different from a second physical computer, on which a second node included in the virtual group and having a degree of similarity to the first node that exceeds the predetermined threshold, operate, moving the first node from the first physical computer to the second physical computer.

10. The operation management method of claim 9 , wherein, communication information between a plurality of nodes includes a communication event count of mutual communication performed between the plurality of nodes.

11. The operation management method of claim 9 , wherein, the operation information indicates whether or not a subset of nodes included in the virtual group operates on a different physical computer from nodes included in the virtual group other than the subset of nodes.

12. The operation management method of claim 9 , further comprising:
adding a node to the virtual group that includes nodes whose degree of similarity to the added node exceeds the specific threshold.

13. The operation management method of claim 12 , wherein as the degree of similarity, a Euclidean distance is found between the added node and one node out of the plurality of nodes.

14. The operation management method of claim 9 , further comprising storing, via a storage section, a virtual group information and the operation information as a table, the virtual group information indicating which subset of the plurality of nodes form a common virtual group.

15. The operation management method of claim 12 , further comprising:
based on the virtual group of the added node and the operation information of the added node, when a physical computer on which the added node operates does not match a physical computer on which a node exhibiting a degree of similarity of the specific value or greater operates, moving the added node to the physical computer on which the node exhibiting a degree of similarity of the specific value or greater operates.

16. The operation management method of claim 9 , further comprising:
when a physical computer including at least one node is in a critical state that is problematic for operation, moving a node that operates on the critical state computer to a physical computer that is different from the critical state computer and that is associated with the operation information of the virtual group including the node.

17. A non-transitory recording medium storing an

Assignees

• Fujitsu Ltd

Inventors

• Takagi Hiroyuki
• Tamura Naohiro

Classifications

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• H04L41/0893

  Assignment of logical groups to network elements · CPC title

• H04L41/00

  Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks · CPC title

• G06F9/5077 Primary

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• H04L43/0817

  by checking functioning · CPC title