System and method for migrating virtual machines on cluster systems

• US 7,730,486 B2
• Filed: 02/28/2005
• Issued: 06/01/2010
• Est. Priority Date: 02/28/2005
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Herington, Daniel E.
• Primary Examiner(s): An; Meng-Ai
• Assistant Examiner(s): To; Jennifer N
• Application Number:
  US11/067,838
• Publication Number:
  US 20060195715A1
• Time in Patent Office: 1,919 Days
• Field of Search: 718/1, 718/104, 709/220, 709223-226
• US Class Current: 718/1
• CPC Class Codes: G06F 11/3433 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 Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455

Abstract

In one embodiment, a method comprises executing a plurality of virtual machines on a plurality of nodes of a cluster computing system, wherein at least one application is executed within each of the plurality of virtual machines, generating data that is related to performance of applications in the virtual machines, analyzing, by a management process, the data in view of parameters that encode desired performance levels of applications, and migrating, by the management process, a virtual machine on a first node to a second node of the plurality of nodes in response to the analyzing.

Claims

1. A method comprising:

executing a plurality of virtual machines on a plurality of nodes of a cluster computing system, wherein at least one application is executed within each of said plurality of virtual machines;

generating data that is related to performance of said applications in said virtual machines;

analyzing, by a management process, said data in view of parameters that encode desired performance levels of said applications;

migrating, by said management process, a virtual machine on a first node to a second node of said plurality of nodes in response to said analyzing; and

wherein said analyzing further comprises determining whether resources are available on said first node for allocation to said virtual machine before performing said migrating, and changing relative entitlement parameters associated with said virtual machine and at least one other virtual machine on said first node in response to said analyzing.

2. The method of claim 1 wherein said cluster computing system comprises a distributed file system that presents a single image of storage resources of said cluster computing system with a single root.

3. The method of claim 2 wherein said virtual machine is assigned virtual storage resources that use storage resources directly attached to said first node.

4. The method of claim 3 wherein said virtual machine continues to access virtual storage resources that use storage resources directly attached to said first node after migration to said second node.

5. The method of claim 2 further comprising:

defining a configuration file, for each of said plurality of virtual machines, that is independent from which node of said plurality of nodes is used to execute the respective virtual machine.

6. The method of claim 1 wherein said determining comprises:

identifying underutilized resources.

7. The method of claim 1 wherein said determining comprises:

identifying unassigned resources.

8. A cluster system comprising:

a plurality of cluster nodes that provide hardware elements to support software operations;

a plurality of virtual machines executing on said plurality of cluster nodes, wherein at least one application is executed within each of said plurality of virtual machines;

a management process located on at least one of said plurality of cluster nodes, wherein said management process analyzes performance data associated with said applications executing within said plurality of virtual machines, and migrates virtual machines between said plurality of cluster nodes in response to analysis of said performance data;

said management process compares said performance data to a plurality of parameters that encode desired operating goals of said applications executed within said plurality of virtual machines; and

said management process determines whether resources are available for reallocation on cluster node executing a virtual machine before migrating said virtual machine to another cluster node, and changes relative entitlement parameters associated with said virtual machine and at least one other virtual machine on said cluster node in response to analyzing said performance data.

9. The cluster system of claim 8 comprising:

a distributed file system implemented within operating systems of said plurality of cluster nodes.

10. The cluster system of claim 9 wherein said distributed file system presents a unified image of storage resources of said cluster system to software processes that is independent from cluster nodes upon which said software processes are executed.

11. The cluster system of claim 9 wherein said distributed file system enables an application, that accesses a virtual disk implemented using a storage peripheral directly attached to a cluster node executing said application, to continue accessing said virtual disk after said application'"'"'s virtual machine is migrated between cluster nodes.

12. The cluster system of claim 8 wherein a respective performance monitor process is executed within each virtual machine to generate said performance data analyzed by said management process.

13. The cluster system of claim 8 further comprising:

a plurality of operating systems executed on said plurality of cluster nodes that virtualize physical resources to implement said plurality of virtual machines.

14. A computer readable medium having code stored thereon for implementing management of applications executed within virtual machines of a cluster system, said code comprising:

code for generating performance data related to execution of said applications within said virtual machines of said cluster system;

code for processing said performance data to determine whether said applications are satisfying desired operating goals encoded within a set of parameters;

code for determining whether resources are available on a single cluster node for allocation to said virtual machines before performing migration;

code for changing relative entitlement parameters associated with said virtual machines on said single cluster node in response to said code for determining; and

code for migrating virtual machines between cluster nodes of said cluster system in response to said code for processing.

15. The computer readable medium of claim 14 said code stored thereon further comprising:

code for reallocating resources between virtual machines on the single cluster node.

16. The computer readable medium of claim 15 wherein said code for migrating is operable when resources not available for reallocation by said code for reallocating.

17. The computer readable medium of claim 14 said code stored thereon further comprising:

code for providing a cluster file system that enables access to storage resources of said cluster system by software processes independently of which cluster node executes said software processes.