Method and apparatus for maintaining a workload service level on a converged platform

Patent metadata

Field	Value
Publication number	US-9348724-B2
Application number	US-201213476895-A
Country	US
Kind code	B2
Filing date	May 21, 2012
Priority date	May 21, 2012
Publication date	May 24, 2016
Grant date	May 24, 2016

Abstract

A management server is provided in a computer system having one or more hosts, one or more storage systems and one or more switches, the hosts having a plurality of virtual machines, each virtual machine being defined according to a service level agreement. The management server is operable to manage the virtual machines and resources associated with the virtual machines; receive a notification of an event from a node in the computer system; determine if the event affects a service level agreement for any of the virtual machines defined in the computer system, the service level agreements listing required attributes for the corresponding virtual machines; allocate a new resource for a virtual machine whose service level agreement is affected by the event; and move the virtual machine whose service level agreement is affected by the event to the newly allocated resource.

First claim

What is claimed is:

1. A management server in a computer system, the computer system including one or more hosts, one or more storage systems, one or more network switches, and a plurality of resources provided using one or more of the storage systems and the network switches, the hosts having a plurality of virtual machines, each virtual machine being defined according to a service level agreement, and each storage system having a storage controller and a plurality of storage devices, the management server comprising:
a processor;
a network interface; and
a memory,
wherein the management server is configured to:
manage, by maintaining and using a plurality of tables of relationships between the virtual machines and resources of the plurality of resources, the virtual machines and resources associated with the virtual machines,
receive a notification of an event from any one of the storage systems and network switches of the computer system, wherein the event indicates a problematic resource having one or more of a fault, an unavailability, an integrity issue, and a performance issue, the problematic resource being a resource of the plurality of resources,
determine an affected virtual machine of the plurality of virtual machines using the indication of the problematic resource and the tables of relationships, the affected virtual machine being affected by the event,
determine to move the affected virtual machine when the event breaks a service level agreement, the service level agreement including required attributes for the affected virtual machine,
allocate a new resource for the affected virtual machine, and
move the affected virtual machine to the newly allocated resource,
wherein the plurality of tables of relationships include:
a first table indicating relationships between the virtual machines and thin provisioned volumes,
a second table indicating relationships between the thin provisioned volumes, pools associated with the thin provisioned volumes, and the storage systems,
a third table indicating relationships between array groups and pool volumes, and
a fourth table indicating relationships between the pool volumes and the pools,

wherein determining the affected virtual machine includes:
when the event indicates a failure of a storage media in an array group, determining using the third table that a pool volume uses the array group, determining using the fourth table that a pool uses the pool volume, determining using the second table that a thin provisioned volume uses the pool, and determining using the first table that the affected virtual machine uses the thin provisioned volume.

2. The management server of claim 1 , wherein the network interface couples the management server to the hosts and the storage systems via a first network, and the hosts and the storage systems are coupled via a second network, wherein the first network is an out-of-band network, and the second network is an in-band network, and wherein management server receives the notification of the event through the out-of-band network.

3. The management server of claim 1 , wherein the management server is configured to manage the service level agreements for the virtual machines, each service level agreement including one or more of the following attributes: a number of nodes included in a cluster associated with the virtual machine, a number of parity disks associated with the virtual machine, a number of paths from a host to a thin provisioned volume, a tier on which a pool is located, a number of cores reserved for the virtual machine.

4. The management server of claim 1 , wherein the computer system includes a converged storage system and an external storage system coupled thereto, the converged storage system including at least one host, at least one network switch, and at least one storage system housed in a single housing.

5. The management server of claim 1 , wherein the management server is operable to:
select a predefined resource for the affected virtual machine when the predefined resource satisfies the service level agreement, and
move the affected virtual machine to the predefined resource,
wherein the new resource is allocated if there is no predefined resource that satisfies the service level agreement.

6. The management server of claim 1 , wherein the management server is operable to:
send a notification to a virtual machine administrator to obtain an approval of the newly allocated resource for the affected virtual machine.

7. The management server of claim 1 , wherein the plurality of tables of relationships includes information indicating:
relationships between the network switches and the storage systems;
relationships between the virtual machines and the hosts;
relationships between the hosts and interfaces of a first plurality of interfaces;
relationships between the thin provisioned volumes and interfaces of a second plurality of interfaces;
relationships between the hosts and cores of a plurality of cores;
relationships between the storage systems and a plurality of storage pools; and
relationships between the resources and attributes of the service level agreement.

8. The management server of claim 1 ,
wherein determining the affected virtual machine further includes:
when the event indicates that a free capacity of the pool is less than a predetermined threshold, determining using the second table that the thin provisioned volume uses the pool, and determining using the first table that the affected virtual machine uses the thin provisioned volume.

9. The management server of claim 1 ,
wherein the plurality of tables of relationships further include:
a fifth table indicating relationships between networks and network interfaces of the hosts, and
a sixth table indicating relationships between the thin provisioned volumes and the network interfaces of the hosts; and

wherein determining the affected virtual machine further includes:
when the event indicates a network failure, determining that an affected network is affected by the event, determining using the fifth table that a network interface of a host uses the affected network, determining using the sixth table that the thin provisioned volume is connected to the network interface, and determining using the first table that the affected virtual machine uses the thin provisioned volume.

10. A method for managing service level agreements of virtual machines in a computer system, the computer system including one or more hosts, one or more storage systems, one or more network switches, and a plurality of resources provided using one or more of the storage systems and the network switches, the virtual machines running in the hosts, each storage system having a storage controller and a plurality of storage devices, the method comprising:
maintaining a plurality of tables of relationships between the virtual machines and resources of the plurality of resources;
receiving a notification of an event, wherein the event indicates a problematic resource, the problematic resource being a resource of the plurality of resources;
determining, using the indication of the problematic resource and the tables of relationships, an affected virtual machine of the virtual machines in the computer system, the affected virtual machine being affected by the event;
determining to move the affected virtual machine when the event breaks a service level agreement, the service level agreement including required attributes for the affected virtual machine;
allocating a new resource for the affected virtual machine; and
moving the affected virtual machine to the newly allocated resource,
wherein the plurality of tables of relations

Assignees

• Ota Hiromichi
• Kano Yoshiki
• Murase Atsushi
• Hitachi Ltd

Inventors

• Ota Hiromichi
• Kano Yoshiki
• Murase Atsushi

Classifications

• G06F11/0775

  Content or structure details of the error report, e.g. specific table structure, specific error fields · CPC title

• G06F11/3485 Primary

  for I/O devices · CPC title

• G06F9/5088

  involving task migration · CPC title

• G06F2009/4557

  Distribution of virtual machine instances; Migration and load balancing · CPC title

• G06F11/0712

  in a virtual computing platform, e.g. logically partitioned systems · CPC title