Exploiting probabilistic latency expressions for placing cloud applications

1. A method for placing components of a plurality of instances of a cloud application on nodes in a cloud infrastructure, the method comprising:

receiving by a cloud controller scenario probabilities of a plurality of cloud application scenarios, wherein the plurality of cloud application scenarios define which of a plurality of application components to use and an order of operation for the plurality of application components;

receiving by a cloud controller cloud infrastructure performance data;

defining a performance metric of the cloud application based upon cloud infrastructure performance data;

defining constraints on the placement of the plurality of application components;

receiving a service level agreement performance requirement based upon a probability of meeting a specified performance requirement;

optimizing the placement of the plurality of application components of the plurality of instances of the cloud application on nodes in the cloud infrastructure based upon the scenario probabilities, cloud infrastructure performance data, the performance metric, and the constraints on the placement of the plurality of application components to meet the service level agreement performance requirement.

2. The method of claim 1 , wherein performance metric is an end-to-end latency of the cloud application and the service level agreement performance requirement is an application response time.

3. The method of claim 2 , wherein end-to-end latency of the cloud application includes a message transmission latency.

4. The method of claim 2 , wherein end-to-end latency of the cloud application includes a computation latency.

5. The method of claim 2 , wherein end-to-end latency of the cloud application includes a storage latency.

6. The method of claim 1 , wherein the cloud infrastructure performance data includes a network latency matrix defining the latency between each available node in the cloud computing infrastructure.

7. The method of claim 1 , wherein the constraints on the placement of the application components includes one of computing constraints, bandwidth constraints, and storage constraints.

8. The method of claim 1 , further comprising receiving computation times and probabilities for a cloud computing application component wherein the optimizing the placement of the components is further based upon the computation times and probabilities for the cloud computing application.

9. A cloud controller method for placing components of a plurality of instances of a cloud application on nodes in a cloud infrastructure the cloud controller comprising:

a network interface;

a memory; and

a processor in communication with the memory, the processor being configured to:

receive scenario probabilities of a plurality of cloud application scenarios, wherein the plurality of cloud application scenarios define which of a plurality of application components to use and an order of operation of the plurality for application components;

define a performance metric of the cloud application based upon cloud infrastructure performance data;

define constraints on the placement of the plurality of application components;

receive a service level agreement performance requirement based upon a probability of meeting a specified performance requirement;

optimize the placement of the components of the plurality of instances of the cloud application on nodes in the cloud infrastructure based upon the scenario probabilities, cloud infrastructure performance data, the performance metric, and the constraints on the placement of the plurality of application components to meet the service level agreement performance requirement.

10. The cloud controller of claim 9 , wherein performance metric is an end-to-end latency of the cloud application and the service level agreement performance requirement is an application response time.

11. The cloud controller of claim 10 , wherein end-to-end latency of the cloud application includes one of a message transmission latency, a computation latency, and a storage latency.

12. The cloud controller of claim 9 , wherein the cloud infrastructure performance data includes a network latency matrix defining the latency between each available node in the cloud computing infrastructure.

13. The cloud controller of claim 9 , wherein the constraints on the placement of the application components includes one of computing constraints, bandwidth constraints, and storage constraints.

14. A non-transitory machine-readable storage medium encoded with instructions for execution by a cloud controller for placing components of a plurality of instances of a cloud application on nodes in a cloud infrastructure, the medium comprising:

instructions for receiving scenario probabilities of a plurality of cloud application scenarios, wherein the plurality of cloud application scenarios define which of a plurality of application components to use and an order of operation for the plurality of application components;

instructions for receiving cloud infrastructure performance data;

instructions for defining a performance metric of the cloud application based upon cloud infrastructure performance data;

instructions for defining constraints on the placement of the plurality of application components;

receiving a service level agreement performance requirement based upon a probability of meeting a specified performance requirement;

instructions for optimizing the placement of the components of the plurality of instances of the cloud application on nodes in the cloud infrastructure based upon the scenario probabilities, cloud infrastructure performance data, the performance metric, and the constraints on the placement of the plurality of application components to meet the service level agreement performance requirement.

15. The non-transitory machine-readable storage medium of claim 14 , wherein performance metric is an end-to-end latency of the cloud application and the service level agreement performance requirement is an application response time.

16. The non-transitory machine-readable storage medium of claim 15 , wherein end-to-end latency of the cloud application includes a message transmission latency.

17. The non-transitory machine-readable storage medium of claim 15 , wherein end-to-end latency of the cloud application includes a computation latency.

18. The non-transitory machine-readable storage medium of claim 15 , wherein end-to-end latency of the cloud application includes a storage latency.

19. The non-transitory machine-readable storage medium of claim 14 , wherein the cloud infrastructure performance data includes a network latency matrix defining the latency between each available node in the cloud computing infrastructure.

20. The non-transitory machine-readable storage medium of claim 14 , wherein the constraints on the placement of the application components includes one of computing constraints, bandwidth constraints, and storage constraints.