System management method, management computer, and non-transitory computer-readable storage medium

Patent metadata

Field	Value
Publication number	US-2016188373-A1
Application number	US-201414909770-A
Country	US
Kind code	A1
Filing date	Feb 12, 2014
Priority date	Feb 12, 2014
Publication date	Jun 30, 2016
Grant date	—

Abstract

A system management method for a management computer coupled to a computer system, the computer system including a plurality of computers, an operations system being built thereon the computer system, the operations system including a plurality of task nodes each having allocated thereto computer resources, the system management method including: a step of analyzing a configuration of the computer system for specifying a important node, which is an important task node in the operations system; a step of changing an allocation amount of the computer resources allocated to the important node for measuring a load of the operations system; a step of calculating a first weighting representing a strength of associations among the plurality of task nodes based on a measurement result of the load; and a step of specifying a range impacted by a change in the load of the important node based on the calculated first weighting.

First claim

What is claimed is:

1 . A system management method for a management computer coupled to a computer system,
the management computer including: a first processor; a first memory coupled to the first processor; and a first interface coupled to the first processor, the computer system including a plurality of computers, each of the plurality of computers including: a second processor; a second memory coupled to the second processor; and a second interface coupled to the second processor, an operations system being built thereon the computer system, the operations system including a plurality of task nodes each having allocated thereto one of computer resources of one computer among the plurality of computers and computer resources of a virtual computer generated on at least one computer among the plurality of computers, the system management method including: a first step of analyzing, by the management computer, a configuration of the computer system for specifying at least one important node, which is an important task node in the operations system; a second step of changing, by the management computer, an allocation amount of the computer resources allocated to the at least one important node for measuring a load of the operations system; a third step of calculating, by the management computer, a first weighting representing a strength of associations among the plurality of task nodes based on a measurement result of the load; and a fourth step of specifying, by the management computer, a range impacted by a change in the load of the at least one important node based on the calculated first weighting.

2 . The system management method according to claim 1 ,
wherein the management computer stores: topology management information for managing an apparatus configuration of each of the plurality of computers and a coupling configuration of the plurality of computers; logical configuration management information for managing an apparatus configuration of a plurality of virtual computers and a coupling configuration of the plurality of virtual computers; and node management information for managing a connection configuration of the plurality of task nodes, and, a correspondence relationship of the one computer and the virtual computer allocating the computer resources to each of the plurality of task nodes, wherein the topology management information includes a first eigen value representing an importance level of a physical configuration in the computer system, the first eigen value being set based on the apparatus configuration of each of the plurality of computers and the coupling configuration of each of the plurality of computers, wherein the logical configuration management information includes a second eigen value representing an importance level of a logical configuration in the computer system, the second eigen value being set based on the apparatus configuration of the plurality of virtual computers and the coupling relationship of each of the plurality of virtual computers, and wherein the first step includes: referring to the node management information for specifying, for each of the plurality of task nodes, the one computer and the virtual computer allocating the computer resources to the each of the plurality of task nodes; referring, for a task node to be allocated with the computer resources of the one computer, to the topology management information for obtaining the at least one first eigen value associated with the one computer allocating the computer resources to the task node, and calculating a second weighting representing an importance level of the task node to be allocated with the computer resources of the one computer based on the at least one first eigen value; referring, for a task node to be allocated with the computer resources of the virtual computer, to the logical configuration management information for obtaining the at least one second eigen value associated with the virtual computer allocating the computer resources to the task node, referring to the logical configuration management information and the topology management information for obtaining the at least one first eigen value associated with a computer generated by the virtual computer allocating the computer resources to the task node, and calculating the second weighting based on the at least one first eigen value and the at least one second eigen value; and specifying at least one of the important nodes from among the plurality of task nodes based on the second weighting of each of the plurality of task nodes.

3 . The system management method according to claim 2 ,
wherein the second step includes: selecting one important node to be processed from among the at least one important node; changing the allocation amount of the computer resources to be allocated to the important node to be processed for measuring the load of each of the plurality of task nodes; and calculating, based on a measurement result of the load, a performance characteristic function representing an association between the allocation amount of the computer resources to be allocated to the important node to be processed and the load of each of the plurality of task nodes, and wherein the third step includes generating an association matrix including the first weighting as an off-diagonal component value and the second weighting as a diagonal component value.

4 . The system management method according to claim 3 ,
wherein each of the plurality of task nodes is configured to execute a predetermined task, wherein the management computer stores task management information in which a type of task to be executed by the operations system, a configuration required for the task, and a fourth eigen value representing an importance level of the task are associated with one another, and wherein the fourth step includes: referring to the node management information for selecting one task node to be processed; referring to the node management information for specifying a connected node, which is a task node connected to the task node to be processed; obtaining, based on the node management information and the task management information, the fourth eigen value corresponding to the type of the task to be executed by the task node to be processed and the fourth eigen value corresponding to the type of the task to be executed by the connected node; calculating a second weighting representing a strength of an association between the task node to be processed and the connected node by using the fourth eigen value of the task node to be processed and the fourth eigen value of the connected node; generating a superposition matrix including the second weighting as an off-diagonal component value and a eigen value of the task node as a diagonal component value; and correcting the association matrix by using the superposition matrix.

5 . The system management method according to claim 3 ,
wherein each of the plurality of task nodes is configured to execute a predetermined task, wherein the management computer stores rule management information in which a type of the task and a method of changing the computer resources to be allocated to the task node executing the task are associated with one another, and wherein the system management method further includes: monitoring, by the management computer, a load of each of the plurality of task nodes included in the impacted range; estimating, by the management computer, a critical value representing a possibility of the computer resources of each of the plurality of task nodes being insufficient by using a value of the load of each of the plurality of task nodes and the performance characteristic function; determining, by the management computer, for each of the plurality

Assignees

• Hitachi Ltd

Inventors

• Tameshige Takashi
• Kudou Yutaka
• Morimura Tomohiro
• Teramura Takeshi
• Iizuka Daisuke
• Ozaki Nobuaki

Classifications

• G06F2201/815

  Virtual · CPC title

• G06F9/5083

  Techniques for rebalancing the load in a distributed system · CPC title

• G06F11/3409

  for performance assessment · CPC title

• G06F9/5077

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

• G06F11/3433 Primary

  for load management (allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029) · CPC title