Automated virtual machine placement planning using different placement solutions at different hierarchical tree levels

• US 8,959,523 B2
• Filed: 03/30/2012
• Issued: 02/17/2015
• Est. Priority Date: 03/30/2012
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Patil, Priti P., Neogi, Anindya, Sen, Aritra, Verma, Akshat
• Primary Examiner(s): An, Meng
• Assistant Examiner(s): Zhao, Bing
• Application Number:
  US13/434,961
• Publication Number:
  US 20130263120A1
• Time in Patent Office: 1,054 Days
• Field of Search: None
• US Class Current: 718/104
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2009/45575 Starting, stopping, suspending or resuming virtual machine instances
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5066 Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs mappping at compile time, see G06F8/451

Abstract

A virtual machine placement framework is described to enable a data center operator to develop a placement scheme to satisfy its particular constraints while simultaneously optimizing resource utilization. To generate a placement solution, the virtual machine placement problem is first characterized as a “bin packing” problem. The framework provides simple interface tools and processing modules, and a pluggable architecture for receiving placement algorithms. To generate a solution, an administrator creates an XML representation that abstracts physical entities (e.g., data center, subnet, rack, physical server, and the like) into a hierarchical tree of bins. The administrator also defines a set of “rules” that govern (direct) the placement of the virtual machines by placing constraints on the placement scheme. Using the hierarchical tree and the rules, the framework is executed to generate a placement as a solution to a bin packing problem, preferably on a layer-by-layer basis.

Claims

1. A method of placing virtual machines, comprising:

representing physical entities upon which the virtual machines are to be placed as a hierarchical tree of bins;

receiving one or more virtual machine-to-physical entity constraints on a placement scheme for the virtual machines; and

using information derived from the hierarchical tree and the virtual machine-to-physical entity constraints to define, for each level of the hierarchical tree, a placement solution that associates given virtual machines to given physical entities by solving, within a computing machine having a hardware element, a bin packing problem, wherein the bin packing problem is solved at a given level of the hierarchical tree using a placement algorithm, together with a placement solution from a prior level.

2. The method as described in claim 1 wherein the bin packing problem is solved at each level of the hierarchical tree using virtual machine-to-physical entity constraints associated with that level and the placement algorithm.

3. The method as described in claim 2 wherein the virtual machine-to-physical entity constraints associated with a first level differ from the virtual machine-to-physical entity constraints associated with a second level.

4. The method as described in claim 2 wherein the placement algorithm used at a first level differs from the placement algorithm used at a second level.

5. The method as described in claim 1 wherein the virtual machine-to-physical entity constraints are defined in one or more user-defined rules.

6. The method as described in claim 1 wherein the using step further comprises:

applying the virtual machine-to-physical entity constraints to a set of virtual machines to partition the set of virtual machines into a set of virtual machine groupings;

for each level in the hierarchical tree, beginning with a first level and continuing until all levels in the hierarchy have been processed;

generating from the set of virtual machine groupings one or more entities;

applying the placement algorithm to the one or more entities to compute the placement solution that associates given virtual machines to given physical entities.

7. The method as described in claim 1 wherein the hierarchical tree is defined as an XML document that models a facility as a tree of physical entities and attributes associated therewith.

21. The computer program product as described in claim 5 wherein the hierarchical tree is defined as an XML document that models a facility as a tree of physical entities and attributes associated therewith.

8. Apparatus, comprising:

a processor;

computer memory holding computer program instructions executed by the processor to place virtual machines, the computer program instructions comprising;

program code to represent physical entities upon which the virtual machines are to be placed as a hierarchical tree of bins;

program code to receive one or more virtual machine-to-physical entity constraints on a placement scheme for the virtual machines; and

program code to use information derived from the hierarchical tree and the virtual machine-to-physical entity constraints to define, for each level of the hierarchical tree, a placement solution that associates given virtual machines to given physical entities by solving a bin packing problem, wherein the bin packing problem is solved at a given level of the hierarchical tree using a placement algorithm, together with a placement solution from a prior level.

9. The apparatus as described in claim 8 wherein the bin packing problem is solved at each level of the hierarchical tree using virtual machine-to-physical entity constraints associated with that level and the placement algorithm.

10. The apparatus as described in claim 9 wherein the virtual machine-to-physical entity constraints associated with a first level differ from the virtual machine-to-physical entity constraints associated with a second level.

11. The apparatus as described in claim 9 wherein the placement algorithm used at a first level differs from the placement algorithm used at a second level.

12. The apparatus as described in claim 8 wherein the virtual machine-to-physical entity constraints are defined in one or more user-defined rules.

13. The apparatus as described in claim 8 wherein the program code to use further comprises:

program code to apply the virtual machine-to-physical entity constraints to a set of virtual machines to partition the set of virtual machines into a set of virtual machine groupings;

for each level in the hierarchical tree, beginning with a first level and continuing until all levels in the hierarchy have been processed;

program code to generate from the set of virtual machine groupings one or more entities;

program code to apply the placement algorithm to the one or more entities to compute the placement solution that associates given virtual machines to given physical entities.

14. The apparatus as described in claim 8 wherein the hierarchical tree is defined as an XML document that models a facility as a tree of physical entities and attributes associated therewith.

15. A computer program product in a non-transitory computer readable storage medium for use in a data processing system, the computer program product holding computer program instructions which, when executed by the data processing system, perform a method of placing virtual machines, the method comprising:

representing physical entities upon which the virtual machines are to be placed as a hierarchical tree of bins;

receiving one or more virtual machine-to-physical entity constraints on a placement scheme for the virtual machines; and

using information derived from the hierarchical tree and the virtual machine-to-physical entity constraints to define, for each level of the hierarchical tree, a placement solution that associates given virtual machines to given physical entities by solving a bin packing problem, wherein the bin packing problem is solved at a given level of the hierarchical tree using a placement algorithm, together with a placement solution from a prior level.

16. The computer program product as described in claim 15 wherein the bin packing problem is solved at each level of the hierarchical tree using virtual machine-to-physical entity constraints associated with that level and the placement algorithm.

17. The computer program product as described in claim 16 wherein the virtual machine-to-physical entity constraints associated with a first level differ from the virtual machine-to-physical entity constraints associated with a second level.

18. The computer program product as described in claim 16 wherein the placement algorithm used at a first level differs from the placement algorithm used at a second level.

19. The computer program product as described in claim 15 wherein the virtual machine-to-physical entity constraints are defined in one or more user-defined rules.

20. The computer program product as described in claim 15 wherein the using step further comprises:

applying the virtual machine-to-physical entity constraints to a set of virtual machines to partition the set of virtual machines into a set of virtual machine groupings;

for each level in the hierarchical tree, beginning with a first level and continuing until all levels in the hierarchy have been processed;

generating from the set of virtual machine groupings one or more entities;

applying the placement algorithm to the one or more entities to compute the placement solution that associates given virtual machines to given physical entities.

22. A virtual machine placement framework, comprising:

a processor;

a configuration tool executed on the processor to generate first data and second data, the first data representing, as a hierarchical tree of data containers, physical entities upon which the virtual machines are to be placed, the second data representing virtual machine-to-physical entity constraints on a placement scheme for the virtual machines; and

a placement engine executed on the processor to receive information derived from the first data and the second data, the placement engine executing one or more placement algorithms to generate a placement solution of virtual machines with respect to the physical entities;

the one or more placement algorithms including a first placement algorithm that is executed at a first level of the hierarchical tree to generate a first placement solution that associates given first virtual machines to given first physical entities, and a second placement algorithm that is executed at a second level of the hierarchical tree to generate a second placement solution that associates given second virtual machines to given second physical entities, the second placement algorithm also using the first placement solution to generate the second placement solution.

23. The virtual machine placement framework as described in claim 22 further including an algorithm selector executed on the processor to select the placement algorithm for use at a given level of the hierarchical tree.

24. The virtual machine placement framework as described in claim 23 wherein the configuration tool includes a rules editor to generate the second data using a rules framework.