Information handling system application decentralized workload management

• US 9,195,510 B2
• Filed: 04/04/2011
• Issued: 11/24/2015
• Est. Priority Date: 04/04/2011
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Hirschfeld, Robert Ari, Althaus, Gregory Scott
• Primary Examiner(s): TANG, KAREN C
• Application Number:
  US13/079,170
• Publication Number:
  US 20120254437A1
• Time in Patent Office: 1,695 Days
• Field of Search: 709/217, 709/224, 709/225
• US Class Current: 1/1
• CPC Class Codes: G06F 9/5044 considering hardware capabilities
• CPC Class Codes: G06F 9/5072 Grid computing
• 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 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

A cloud application management infrastructure models biological swarm behaviors to assign application resources to physical processing resources in a decentralized manner. A balanced and highly automated management of cloud infrastructure has a predictable and reliable response to changing resource loads by using a limited local rule set to define how application instances interact with available resources. Digital pheromone signals at physical resources are applied locally by a swarm module to determine if the physical resources provide an acceptable environment for an application and, if not, the application swarms to other environments until a suitable environment is found.

Claims

1. An information handling system comprising:

plural processing resources interfaced by a network;

plural containers, each container executing on an associated processing resource;

a pheromone list associated with each processing resource; and

plural swarm modules, each container having a single distinct associated swarm module to provide decentralized management of each container'"'"'s assignment to one of the processing resources selected by the container'"'"'s single distinct swarm module, each swarm module assigned to a single distinct container, the swarm module operable to model emergent biologic behavior, including at least to retrieve the pheromone list from its associated processing resource to determine locally at the associated processing resource a selected of the plural resources for executing the container and to write one or more pheromones to the pheromone list;

wherein the pheromone list comprises an allergen, the swarm module responding to the presence of an allergen in the pheromone list by moving the container to a different processing resource.

2. The information handling system of claim 1 wherein the pheromone list comprises an aggregate value, the swarm module responding to an aggregation value of greater than a predetermined value by moving the container to a different processing resource.

3. The information handling system of claim 1 wherein the pheromone list comprises an aggregate value, the swarm module responding to an aggregation value of less than a predetermined value by moving the container to a different processing resource.

4. The information handling system of claim 1 wherein the pheromone list comprises a swarm pheromone associated with a predetermined type of container, the swarm module responding to the lack of a swarm pheromone in the pheromone list by moving the container to a different processing resource.

5. The information handling system of claim 1 wherein the swarm module responds to an aggregate value of the swarm pheromone in the pheromone list of greater than a predetermined valued by moving the container to a different processing resource.

6. The information handling system of claim 1 wherein each container generates one or more predetermined pheromones to add to the pheromone list of the processing resource that executes the container.

7. The information handling system of claim 6 wherein each container comprises plural applications, each application generating one or more predetermined pheromones.

8. The information handling system of claim 6 wherein each container comprises a virtual machine.

9. A method for managing applications in a cloud having plural processing resources interfaced by a network, the method comprising:

listing factors at each processing resource of the containers running on the processing resource;

executing plural swarm modules on the processing resources, each container having a distinct associated swarm module executing on the same processing resource as the container, each swarm module having a single distinct container;

evaluating the factors listed at the processing resource with the distinct swarm module associated with each container, the distinct swarm module running locally on the same processing resource as the container with which the distinct swarm module is associated to provide decentralized management of the resources on which the container associated with the distinct swarm module executes, the distinct swarm module determining locally on the processing resource whether to execute its distinct associated container at the processing resource or to move the distinct associated container through the network to a different processing resource;

wherein a factor comprises an allergen, the method further comprising;

determining the allergen applies to the container; and

moving the container to a different processing resource in response to the determining.

10. The method of claim 9 wherein the factor comprises an aggregate of factors having less than a predetermined value, the method further comprising:

determining the aggregate applies to the container; and

moving the container to a different processing resource in response to the determining.

11. The method of claim 9 wherein the factor comprises an aggregate of factors having greater than a predetermined value, the method further comprising:

determining the aggregate applies to the container; and

moving the container to a different processing resource in response to the determining.

12. The method of claim 9 further comprising:

searching for a swarm factor at the processing resource;

failing to locate the swarm factor at the processing resource; and

responding to the failing by moving the container from the processing resource through the network to another of the processing resources having the swarm factor.

13. The method of claim 9 further comprising:

searching for a swarm factor at the processing resource;

locating the swarm factor at the processing resource; and

responding to the locating by initiating the container at the processing resource.

14. The method of claim 13 further comprising:

determining that the swarm factor has an aggregate greater than a predetermined value; and

responding to the determining by moving the container from the processing resource through the network to a different processing resource.

15. The method of claim 9 further comprising:

executing the container at the processing resource; and

in response to the executing, adding a factor associated with the container to the listing of factors.