Client placement in a computer network system using dynamic weight assignments on resource utilization metrics

Patent metadata

Field	Value
Publication number	US-9298512-B2
Application number	US-201213594808-A
Country	US
Kind code	B2
Filing date	Aug 25, 2012
Priority date	Aug 25, 2012
Publication date	Mar 29, 2016
Grant date	Mar 29, 2016

Abstract

A system and method for placing a client in a computer network system uses continuously variable weights to resource utilization metrics for each candidate device, e.g., a host computer. The weighted resource utilization metrics are used to compute selection scores for various candidate devices to select a target candidate device for placement of the client.

First claim

What is claimed is:

1. A method for placing a client in a computer network system, the method comprising:
collecting a plurality of resource utilization metrics for each candidate device from a group of candidate devices in the computer network system that can support the client;
assigning continuously variable weights to the resource utilization metrics for each candidate device, wherein each of the resource utilization metrics is assigned at least one of the continuously variable weights and wherein each of the continuously variable weights is a function of a corresponding resource utilization metric;
computing a selection score using the resource utilization metrics with the continuously variable weights for each candidate device, wherein the selection score is computed by multiplying one of the resource utilization metrics with one of the continuously variable weights assigned to that resource utilization metric; and
selecting a target candidate device from the group of candidate devices for placement of the client based on the selection score of the target candidate device for resource utilization balancing in the computer network system, wherein the collecting the resource utilization metrics, the assigning the continuously variable weights, the computing the selection score and the selecting the target candidate device are performed using one or more processors.

2. The method of claim 1 , wherein the assigning the continuously variable weights includes assigning the continuously variable weights using at least one non-linear convex function.

3. The method of claim 2 , wherein the assigning the continuously variable weights includes assigning the continuously variable weights using at least the following function:

∝

×

(

1

1
-
x

)

β

,

where α and β are values greater than zero and x is the utilization of a particular resource.

4. The method of claim 1 , wherein the assigning the continuously variable weights includes assigning the continuously variable weights using at least two different functions for two different continuously variable weights.

5. The method of claim 1 , further comprising normalizing the resource utilization metrics prior to the assigning of the continuously variable weights to the resource utilization metrics.

6. The method of claim 1 , wherein the candidate devices include datastores, host computers or clusters of host computers.

7. The method of claim 6 , wherein the client to be placed is a virtual machine.

8. The method of claim 1 , further comprising finding suitable candidate devices that can support the client using static properties of the candidate devices that define capabilities of the candidate devices.

9. The method of claim 1 , wherein the computing the selection score includes summing products of the resource utilization metrics and the continuously variable weights to derive the selecting score.

10. The method of claim 9 , wherein the selecting the target candidate device includes selecting the lowest selection score from all the selection scores of the candidate devices to select the target candidate device.

11. A system comprising:
a plurality of host computers; and
a client placement module operably connected to the plurality of host computers, the client placement module being configured to collect a plurality of resource utilization metrics for each host computer from a group of the host computers that can support a client to be placed in the system, the client placement module comprising:
a dynamic weight adjusting unit configured to assign continuously variable weights to the resource utilization metrics for each host computer from the group of host computers, wherein each of the resource utilization metrics is assigned at least one of the continuously variable weights and wherein each of the continuously variable weights is a function of a corresponding resource utilization metric; and
a selection score computing unit configured to compute a selection score using the resource utilization metrics with the continuously variable weights for each host computer from the group of host computers, wherein the selection score is computed by multiplying one of the resource utilization metrics with one of the continuously variable weights assigned to that resource utilization metric,
wherein the client placement module is configured to select a target host computer from the group of host computers for placement of the client based on the selection score of the target host computer for resource utilization balancing in the system.

12. The system of claim 11 , wherein the dynamic weight adjusting unit is configured to assign the continuously variable weights using at least one non-linear convex function.

13. The system of claim 12 , wherein the dynamic weight adjusting unit is configured to assign the continuously variable weights using at least the following function:

∝

×

(

1

1
-
x

)

β

,

where α and β are values greater than zero and x is the utilization of a particular resource.

14. The system of claim 11 , wherein the dynamic weight adjusting unit is configured to assign the continuously variable weights using at least two different functions for at least two different continuously variable weights.

15. The system of claim 11 , wherein the client placement module further includes a normalization unit configured to normalize the resource utilization metrics prior to assignment of the continuously variable weights to the resource utilization metrics.

16. The system of claim 11 , wherein the candidate devices include datastores, host computers or clusters of host computers.

17. The system of claim 16 , wherein the client to be placed is a virtual machine.

18. The system of claim 11 , wherein the client placement module further includes a suitability filter that is configured to find suitable candidate devices that can support the client using static properties of the candidate devices that define capabilities of the candidate devices.

19. The system of claim 11 , wherein the selection score computing unit is configured to sum products of the resource utilization metrics and the continuously variable weights to derive the selecting score.

20. The system of claim 19 , wherein the client placement module is configured to select the lowest selection score from all the selection scores of the candidate devices to select the target candidate device

Assignees

• Gulati Ajay
• Shanmuganathan Ganesha
• Varman Peter Joseph
• Padala Pradeep
• Kesavan Mukil
• Vmware Inc

Inventors

• Gulati Ajay
• Shanmuganathan Ganesha
• Varman Peter Joseph
• Padala Pradeep
• Kesavan Mukil

Classifications

• G06F9/505 Primary

  considering the load · 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

• G06F9/4856

  resumption being on a different machine, e.g. task migration, virtual machine migration (G06F9/5088 takes precedence) · CPC title