Method for placing composite applications in a federated environment

1. A method for placing at least one composite application in a federated environment, comprising the steps of:

analyzing a composite application to be deployed in a federated environment to determine requirements of said composite application;

obtaining one or more application artifacts;

analyzing feasibility of placing one or more application components at one or more clusters in the federated environment without knowledge of resource availability at each of the one or more clusters; and

generating a mapping of the one or more application artifacts to the one or more clusters such that said composite application requirement is met, wherein the one or more application artifacts are distributed across a federated environment.

2. The method of claim 1 , wherein the mapping meets one or more application objectives and one or more server cluster objectives.

3. The method of claim 2 , wherein a tradeoff between the one or more application objectives and the one or more server cluster objectives is met flexibly.

4. The method of claim 2 , wherein the one or more application objectives comprise minimizing application cost, wherein application cost comprises at least one or a power cost, a network cost and a hosting cost.

5. The method of claim 1 , wherein the at least one application comprises at least one multi-component application.

6. The method of claim 5 , wherein the at least one multi-component application comprise at least one application wherein a number of replicas of each component varies based on a selected server cluster as a candidate solution.

7. The method of claim 1 , wherein the application requirement comprises one or more SLA goals.

8. The method of claim 7 , wherein the one or more SLA goals comprise at least one of a performance model and application availability.

9. The method of claim 8 , wherein application availability is automatically inferred from site and link availability.

10. The method of claim 8 , wherein the performance model is derived using a mixed-methodology, wherein the mixed-methodology comprises using a mix of one or more measurements and expert information, wherein the one or more measurements comprise one or more third-party models that map response time to utilization for one or more classes of applications.

11. The method of claim 1 , wherein the application requirements comprise one or more infrastructure level parameters.

12. The method of claim 1 , further comprising, for each of one or more components of the composite application:

selecting the cheapest solution;

improving a component with a highest increase in service level agreement (SLA) goal per unit increase in cost; and

iterating until each of one or more SLA goals is met.

13. The method of claim 1 , further comprising:

selecting an initial feasible solution with a lowest cost and one or more load imbalances;

computing a quality of each movement;

selecting a movement with a highest improvement; and

repeating the selecting, computing and selecting steps until a desired load factor is met for each of the one or more clusters.

14. An article of manufacture comprising a computer readable medium having computer readable program code which when executed by a computer, places at least one composite application in a federated environment, said article of manufacture including:

computer readable program code for analyzing a composite application to be deployed in a federated environment to determine requirements of said composite application;

computer readable program code for obtaining one or more application artifacts;

computer readable program code for analyzing one or more available resources in the federated environment without knowledge of resource availability at each of one or more individual sites; and

computer readable program code for generating a mapping of the one or more application artifacts to the one or more available resources such that an said composite application requirement is met, wherein the one or more application artifacts are distributed across a federated environment.

15. The article of manufacture of claim 14 , further comprising, for each of one or more components of the composite application:

computer readable program code for selecting the cheapest solution;

computer readable program code for improving a component with a highest increase in service level agreement (SLA) goal per unit increase in cost; and

computer readable program code for iterating until each of one or more SLA goals is met.

16. A system for placing at least one composite application in a federated environment, comprising:

a memory;

at least one processor coupled to said memory and

a computer readable medium having computer readable program code embodied thereon which when loaded into said memory causes said at least one processor to:

analyze a composite application to be deployed in a federated environment to determine requirements of said composite application;

obtain one or more application artifacts;

analyze one or more available resources in the federated environment without knowledge of resource availability at each of one or more individual sites; and

generate a mapping of the one or more application artifacts to the one or more available resources such that said composite application requirement is met, wherein the one or more application artifacts are distributed across a federated environment.

17. The system of claim 16 , wherein the at least one application comprises at least one multi-component application.

18. The system of claim 16 , wherein in generating a mapping, the computer readable medium further comprises computer readable program code embodied thereon which when loaded into said memory causes said at least one processor to generate a mapping, wherein the mapping meets one or more application objectives and one or more server cluster objectives.

19. The system of claim 16 , wherein the computer readable medium further comprises computer readable program code embodied thereon which when loaded into said memory causes said at least one processor to, for each of one or more components of the composite application:

select the cheapest solution;

improve a component with a highest increase in service level agreement (SLA) goal per unit increase in cost; and

iterate until each of one or more SLA goals is met.

20. The system of claim 16 , wherein the computer readable medium further comprises computer readable program code embodied thereon which when loaded into said memory causes said at least one processor to:

select an initial feasible solution with a lowest cost and one or more load imbalances;

compute a quality of each movement;

select a movement with a highest improvement; and

repeat the selecting, computing and selecting steps until a desired load factor is met for each of the one or more clusters.