METHODS AND SYSTEMS OF MANAGING RESOURCES ALLOCATED TO GUEST VIRTUAL MACHINES

• US 20120174097A1
• Filed: 10/24/2011
• Published: 07/05/2012
• Est. Priority Date: 01/04/2011
• Status: Active Grant

• Current Assignee: Host Dynamics Limited
• Original Assignee: Host Dynamics Limited
• Inventors: LEVIN, Alexander
• Granted Patent: US 8,667,496 B2
• Time in Patent Office: Days
• US Class Current: 718/1
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2209/508 Monitor
• 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

A method of managing one or more resources allocated to a plurality of guest virtual machines. The method comprises executing a plurality of guest virtual machines in a host platform managing at least one resource, allocating the at least one resource among the plurality of guest virtual machines, separately collecting, from each the guest virtual machine, resource usage data by locally monitoring at least one of an operating system operation, an executed procedure, and an executed tread in each the guest virtual machine, gathering the resource usage data from each the guest virtual machine, and reallocating the at least one resource among the plurality of guest virtual machines according to an analysis of the gathered resource usage data from the plurality of guest virtual machines.

Claims

1. A method of managing at least one resource allocated to a plurality of guest virtual machines, comprising:

executing a plurality of guest virtual machines in a host platform managing at least one resource;

allocating said at least one resource among said plurality of guest virtual machines;

separately collecting, from each said guest virtual machine, resource usage data by locally monitoring at least one of an operating system operation, an executed procedure, and an executed tread in each said guest virtual machine;

gathering said resource usage data from each said guest virtual machine; and

reallocating said at least one resource among said plurality of guest virtual machines according to an analysis of said gathered resource usage data from said plurality of guest virtual machines.

2. The method of claim 1, wherein said gathering is performed in a plurality of sampling periods, said reallocating comprising performing a statistical analysis of said resource usage data from each said guest virtual machine and reallocating said at least one resource among said plurality of guest virtual machines according to said statistical analysis.

3. The method of claim 1, wherein said reallocating is performed by instructing a basic input output system (BIOS) of at least one of said plurality of guest virtual machines to change local allocation of at least one virtual resource.

4. The method of claim 1, wherein said reallocating is performed by instructing a guest operating system of at least one of said plurality of guest virtual machines to change local allocation of at least one virtual resource.

5. The method of claim 1, wherein said reallocating is performed by instructing at least one driver separately executed by at least one of said plurality of guest virtual machines to change local allocation of at least one virtual resource at said at least one guest virtual machine.

6. The method of claim 1, wherein said resource usage data comprises at least one of a deficiency and an excess of said at least one resource in a respective said guest virtual machine.

7. The method of claim 1, wherein said reallocating is performed without a shutdown or a halt of an operating system which is responsible for the allocation of resources to procedures in any of said plurality of guest virtual machines.

8. The method of claim 1, further comprising monitoring the availability of said at least one resource at said host platform;

wherein said reallocating is performed according to said monitoring.

9. The method of claim 1, wherein said reallocating is performed by sending at least one hardware interrupt to a basic input output system (BIOS) or a driver of at least one of said plurality of guest virtual machines.

10. The method of claim 1, wherein said monitoring comprises monitoring a member of a group consisting of a kernel process, a kernel thread, a scheduling application, and kernel accessible application, said member being executed at least one of said plurality of guest virtual machines.

11. The method of claim 1, wherein said reallocating comprises calculating a cumulative contribution of a plurality of resource allocations according to said analysis, selecting one of said plurality of resource allocations according to said calculating, and performing said reallocating according to said selecting.

12. The method of claim 1, wherein said method is implemented by a module executed on a virtualization layer of said host platform.

13. The method of claim 1, wherein monitoring is performed by monitoring process scheduling of an operating system managing said at least one locally executed procedure in each said guest virtual machine.

14. A host platform for hosting a plurality of guest virtual machines and managing at least one resource allocated thereto, comprising:

a memory for hosting a plurality of guest virtual machines;

a plurality of monitoring modules each installed to collect resource usage data by locally monitoring at least one of an operating system operation, an executed procedure, and an executed tread in one of said plurality of guest virtual machines;

at least one resource; and

a host controller which communicates with each said monitoring module to reallocate said at least one resource among said plurality of guest virtual machines according to an analysis of said collected resource usage data.

15. The host platform of claim 14, wherein each said monitoring module estimates a desired change in the allocating said at least one resource to a respective said hosting guest virtual machine according to said resource usage data and forwards a request to said host controller accordingly.

16. The host platform of claim 14, wherein said plurality of monitoring modules are installed in the user mode level of the operating system of said plurality of guest virtual machines.

17. The host platform of claim 14, wherein each said monitoring module is configured to update the basic input output system (BIOS) of the hosting said guest virtual machine.

18. The host platform of claim 14, wherein each said monitoring module is configured to communicate with at least one driver executed on the hosting said guest virtual machine.

19. The host platform of claim 14, wherein said host controller communicates with each said monitoring module via a connection channel with a guest operating system of a respective of said plurality of guest virtual machines.

20. The host platform of claim 14, wherein said monitors the availability of said at least one resource, and reallocates said at least one resource according to said monitoring.

21. The host platform of claim 14, wherein said at least one resource comprises a member of a group consisting of:

a central processing unit, a random access memory (RAM), a non-uniform memory architecture (NUMA) node, and a peripheral component interconnect (PCI) card.

22. The host platform of claim 14, wherein said host controller calculates a report summarizing statistic resource usage data of at least some of said guest virtual machines based on said analysis.

23. A computer program product, comprising at least one computer usable medium having a computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a method of managing at least one resource allocated to a plurality of guest virtual machines, said method comprising:

executing a plurality of guest virtual machines in a host platform managing at least one resource;

allocating said at least one resource among said plurality of guest virtual machines;

separately collecting, from each said guest virtual machine, resource usage data by locally monitoring at least one of an operating system operation, an executed procedure, and an executed tread in each said guest virtual machine;

gathering said resource usage data from each said guest virtual machine; and

reallocating said at least one resource among said plurality of guest virtual machines according to an analysis of said gathered resource usage data from said plurality of guest virtual machines.