METHODS AND SYSTEMS TO MANAGE COMPUTER RESOURCES IN ELASTIC MULTI-TENANT CLOUD COMPUTING SYSTEMS

• US 20140280956A1
• Filed: 03/14/2013
• Published: 09/18/2014
• Est. Priority Date: 03/14/2013
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Shu, Guoqiang, Shanmuganathan, Ganesha
• Granted Patent: US 9,571,567 B2
• Time in Patent Office: Days
• US Class Current: 709/226
• CPC Class Codes: G06F 2209/5011 Pool
• CPC Class Codes: G06F 2209/5014 Reservation
• CPC Class Codes: G06F 2209/504 Resource capping
• 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
• CPC Class Codes: H04L 41/082 the condition being updates or upgrades of network functionality
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/0895 Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements
• CPC Class Codes: H04L 41/40 using virtualisation of network functions or resources, e.g. SDN or NFV entities
• CPC Class Codes: H04L 41/5096 wherein the managed service relates to distributed or central networked applications
• CPC Class Codes: H04L 67/1004 Server selection for load balancing
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

Methods and systems to reconfigure clusters in elastic multi-tenant cloud computing system. An example method includes partitioning a first resource reservation of a first virtual data center between a first cluster and a second cluster and partitioning a second resource reservation of a second virtual data center between the first cluster and the second cluster, and based on the partitioning of the first resource reservation and the second resource reservation, collectively adjusting a first portion of the first resource reservation allotted to the first cluster and a second portion of the second resource reservation allotted to the first cluster in a same reconfiguration operation.

Claims

1. A method comprising:

partitioning a first resource reservation of a first virtual data center between a first cluster and a second cluster and partitioning a second resource reservation of a second virtual data center between the first cluster and the second cluster; and

based on the partitioning of the first resource reservation and the second resource reservation, collectively adjusting a first portion of the first resource reservation allotted to the first cluster and a second portion of the second resource reservation allotted to the first cluster in a same reconfiguration operation.

2. The method of claim 1 wherein the first portion and the second portion are adjusted based on a first deployment event affecting the first virtual data center and a second deployment event affecting the second virtual data center, respectively, and the reconfiguration operation is delayed relative to a first time at which the first deployment occurs and a second time at which the second deployment occurs.

3. A method as defined in claim 1 wherein the partitioning occurs in response to a trigger signal.

4. A method as defined in claim 2 wherein the trigger signal is a current trigger signal, the first portion and the second portion are adjusted based on a first deployment event affecting the first virtual data center and a second deployment event affecting the second virtual data center, respectively, and the first and second deployment events occur in a time intervening between the current trigger signal and an earlier trigger signal.

5. A method as defined in claim 2 wherein the reconfiguration operation is a first reconfiguration operation, the method further comprising:

upon detecting the first deployment event, determining whether the first cluster includes sufficient memory resources to enable power-on of a virtual machine deployed on the first cluster in the first deployment event; and

when the memory resources of the first cluster are determined to be insufficient, performing a second reconfiguration operation on the first cluster, without delay, to increase the memory resources allotted to the first cluster by an amount sufficient to enable power-on of the virtual machine.

6. A method as defined in claim 4 wherein the current trigger signal occurs when the intervening time reaches or exceeds a threshold amount of time, or a number of deployment events occurring after the earlier trigger signal reaches or exceeds a threshold number of deployment events.

7. A method as defined in claim 4 wherein multiple deployment events affecting the first virtual data center occur in the intervening time and the partitioning of the first resource reservation is based on a net change in a number of virtual machines associated with the multiple deployment events.

8. A method as defined in claim 1, wherein the reconfiguration operation is to include estimating a number of virtual machines to be deployed in the first virtual data center on the first cluster subsequent to the reconfiguration operation, the reconfiguration operation further adjusting the first portion by an amount sufficient to enable power-on of the estimated number of virtual machines.

9. A configuration system comprising:

a partitioner to partition a first resource reservation of a first virtual data center between a first cluster and a second cluster and to partition a second resource reservation of a second virtual data center between the first cluster and the second cluster; and

a resource adjuster to adjust, based on the partitioning of the first resource reservation and the partitioning of the second resource reservation, a first portion of the first resource reservation allotted to the first cluster and a second portion of the second resource reservation allotted to the first cluster in a same reconfiguration operation.

10. A configuration system as defined in claim 10 wherein the resource adjuster is to adjust the first portion and the second portion based on a first deployment event affecting the first virtual data center and a second deployment event affecting the second virtual data center, respectively, and to adjust the first and second portions after both the first deployment and the second deployment events have occurred.

11. A configuration system as defined in claim 10 further comprising a trigger generator to generate a trigger signal, the partitioner partitioning the first and second resource reservations based on the trigger signal.

13. A configuration system as defined in claim 10 wherein the reconfiguration operation is a first reconfiguration operation, and the resource adjuster is a first resource adjuster, the system further comprising:

a detector to detect the first deployment event and to determine whether the first cluster includes sufficient memory resources to enable power-on of a virtual machine deployed on the first cluster in the first deployment event; and

a second resource adjuster to perform, based on the determination, a second reconfiguration operation on the first cluster, without delay, to increase the memory resources allotted to the first cluster by an amount sufficient to enable power-on of the virtual machine.

12. A configuration system as defined in claim 12 wherein the trigger signal is a current trigger signal, the resource adjuster is to adjust the first portion and the second portion based on a first deployment event affecting the first virtual data center and a second deployment event affecting the second virtual data center, respectively, and the first and second deployment events occurring in a time intervening between the trigger signal and an earlier trigger signal.

14. A method as defined in claim 12 wherein the trigger signal is a current trigger signal, the partitioner partitions the first resource reservation based on a first deployment event affecting the first virtual data center and partitions the second resource reservation based on a second deployment event affecting the second virtual data center, and the trigger generator is further to generate an earlier trigger, the first and second deployment events occurring in an amount of time intervening between the current trigger signal and the earlier trigger signal.

16. A configuration system as defined in claim 14 wherein the first resource adjuster and the second resource adjuster are a same resource adjuster.

15. A configuration system as defined in claim 15 wherein multiple deployment events affecting the first virtual data center occur in the intervening time and the partitioner partitions the first resource reservation based on a net change in a number of virtual machines associated with the multiple deployment events.

17. A tangible machine readable medium comprising instructions that, when executed, cause a machine to at least:

partition multiple virtual data center resource reservations between a first cluster and a second cluster to determine a respective portion of each respective resource reservation to be allotted to the first cluster; and

based on the respective portions, collectively adjust respective resources assigned to the respective cluster resource pools of the first cluster in a same reconfiguration operation.

24. A tangible machine readable medium as defined in claim 17 wherein reconfiguration of the first cluster is performed in a first reconfiguration operation, the instructions further causing the machine to:

determine whether a respective one of the resource cluster pools has insufficient memory resources to support power-on of a virtual machine associated with a respective deployment event; and

,

perform a second reconfiguration operation of the first cluster to increase the memory resources of the respective resource cluster pool by an amount sufficient to permit power-on of the virtual machine, the first reconfiguration being performed independently of the second reconfiguration and the second reconfiguration being performed before the threshold amount of time has elapsed.

18. A tangible machine readable medium as defined in claim 18 wherein the respective portions are determined based on respective deployment events and the reconfiguration operation is delayed until after all of the deployment events have occurred.

19. A tangible machine readable medium as defined in claim 18 wherein the partitioning occurs in response to a trigger signal.

21. A tangible machine readable medium as defined in claim 18 wherein the reconfiguration operation is a first reconfiguration operation, the instructions further causing the machine:

in response to detecting a first of the deployment events, determine whether a first of the cluster resource pools has sufficient memory resources to enable power-on of a virtual machine deployed in the first deployment event; and

when the resources of the first cluster resource pool are insufficient, performing a second reconfiguration operation on the first cluster, without delay, to increase the memory resources assigned to the first cluster resource pool by an amount sufficient to enable power-on of the virtual machine.

22. A tangible machine readable medium as defined in claim 21 wherein the current trigger signal occurs when the intervening time reaches or exceeds a threshold amount of time, or a number of deployment events occurring after the earlier trigger signal reaches or exceeds a threshold number of deployment events.

23. A tangible machine readable medium as defined in claim 21 wherein a first portion of a first virtual data center resource reservation is determined based on a net change in a number of virtual machines deployed in a first virtual data center between the current trigger signal and the earlier trigger signal.

20. A tangible machine readable medium as defined in claim 20 wherein the trigger signal is a current trigger signal, the respective portions are determined based on respective deployment events, and the respective deployment events occur in time intervening between the current trigger signal and an earlier trigger signal.