Virtualization and consolidation analysis engine for enterprise data centers

1. A method for consolidating a plurality of applications into one or more servers, the method comprising:

organizing consolidation constraints in a computer process, the consolidation constraints representing preferences about placing applications into the one or more servers; and

allocating the applications into the one or more servers in a computer process that maximally satisfies the consolidation constraints;

wherein the organizing of the consolidation constraints includes:

representing each of the consolidation constraints according to dependencies between consolidated applications, the dependencies being represented by application affinity matrices, each entry having a value representing the suitability of placing two applications selected from the plurality of applications on a same one of the one or more servers, the suitability controlled by a given one of the consolidation constraints, the values including positive values representing a preference for placing the two applications on the same server and negative values representing an opposition to placing the two applications on the same server, and

classifying the application affinity matrices into a plurality of different classes each of the classes based on the strictness of the corresponding constraint.

2. The method according to claim 1 , wherein the organizing of the consolidation constraints includes discovering the consolidation quality factors.

3. The method according to claim 2 , wherein at least one of the consolidation constraints is discovered from resource utilization traces of the applications.

4. The method according to claim 1 , wherein the application affinity matrices of one class are defined by an entry value which indicates that two of the applications must be placed together on a server.

5. The method according to claim 1 , wherein the application affinity matrices of one class are defined by an entry value which indicates that two of the applications should not be placed together on a server.

6. The method according to claim 1 , wherein the application affinity matrices of one class are defined by an entry value which indicates that two of the applications can be placed together on a server.

7. The method according to claim 1 , wherein the organizing of the consolidation constraints includes combining similar classes of application affinity matrices to reduce the number of application affinity matrices.

8. The method according to claim 1 , wherein the allocating of the applications into the one or more servers is performed with an evolution-based process.

9. The method according to claim 1 , further comprising grouping the applications into a plurality of groups prior to the allocating of the applications into the one or more servers.

10. The method according to claim 1 , wherein one or more of the applications is encapsulated in a virtual machine followed by an encoding scheme to represent the placement of virtual machine across physical machines.

11. An apparatus for consolidating a plurality of applications into one or more servers, the apparatus comprising:

a processor executing instructions for:

organizing consolidation constraints representing preferences about placing applications into the one or more servers; and

allocating the applications into the one or more servers in a manner that maximally satisfies the consolidation constraints,

wherein the organizing of the consolidation constraints includes:

representing each of the consolidation constraints according to dependencies between consolidated applications, the dependencies being represented by application affinity matrices, each entry having a value representing the suitability of placing two applications selected from the plurality of applications on a same one of the one or more servers, the suitability controlled by a given one of the consolidation constraints, the values including positive values representing a preference for placing the two applications on the same server and negative values representing an opposition to placing the two applications on the same server, and

classifying the application affinity matrices into a plurality of different classes each of the classes based on the strictness of the corresponding constraint.

12. The apparatus according to claim 11 , wherein the organizing of the consolidation constraints includes discovering the consolidation quality factors.

13. The apparatus according to claim 12 , wherein at least one of the consolidation constraints is discovered from resource utilization traces of the applications.

14. The apparatus according to claim 11 , wherein the application affinity matrices of one class are defined by an entry value which indicates that two of the applications must be placed together on a server.

15. The apparatus according to claim 11 , wherein the application affinity matrices of one class are defined by an entry value which indicates that two of the applications should not be placed together on a server.

16. The apparatus according to claim 11 , wherein the application affinity matrices of one class are defined by an entry value which indicates that two of the applications can be placed together on a server.

17. The apparatus according to claim 11 , wherein the organizing of the consolidation constraints includes combining similar classes of application affinity matrices to reduce the number of application affinity matrices.

18. The apparatus according to claim 11 , wherein the allocating of the applications into the one or more servers is performed with an evolution-based process.

19. The apparatus according to claim 11 , wherein the instructions for grouping the applications into a plurality of groups prior to the allocating of the applications into the one or more servers.