Method and system for utilizing spare cloud resources

• US 9,471,384 B2
• Filed: 03/16/2012
• Issued: 10/18/2016
• Est. Priority Date: 03/16/2012
• Status: Active Grant

• Current Assignee: Rackspace US Incorporated (Apollo Global Management, Inc.)
• Original Assignee: Rackspace US Incorporated (Apollo Global Management, Inc.)
• Inventors: Messerli, Antony Joel
• Primary Examiner(s): Tang, Kenneth
• Application Number:
  US13/422,135
• Publication Number:
  US 20130247034A1
• Time in Patent Office: 1,677 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2009/45595 Network integration; Enabling network access in virtual machine instances
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5038 considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration scheduling strategies G06F9/4881 and subgroups
• CPC Class Codes: G06F 9/505 considering the load

Abstract

A cloud computing system including a computing device configured to run virtual machine instances is disclosed. The computing device includes a hypervisor program for managing the virtual machine instances. A customer virtual machine instance is run by the hypervisor program on the computing device, and a grid virtual machine instance is run by the hypervisor program on the computing device. The grid virtual machine instance is configured to run only when a resource of the computing device is not being utilized by the customer virtual machine instance.

Claims

1. A method for utilizing spare resources in a cloud computing system including a plurality of computing devices, the method comprising:

running a grid virtual machine instance on each of the plurality of computing devices;

receiving a work request via a first application programming interface (API);

determining a composite load factor on each of the plurality of computing devices, wherein the composite load factor represents the overall load on the respective computing device;

examining the composite load factor on each of the plurality of computing devices to produce a first set of computing devices with composite load factors below a threshold;

dividing the work request into a plurality of subtasks; and

instructing, via a second API, the grid virtual machine instance running on each computing device in the first set of computing devices to process one of the plurality of subtasks.

2. The method of claim 1, further comprising:

examining the customer load factor of each of the plurality of computing devices to produce a second set of computing devices with customer load factors below the threshold;

for each of the plurality of computing devices that is a member of the first set and not the second set, instructing the grid virtual machine instance running on the computing device to stop processing the subtask it was previously instructed to process; and

for each of the plurality of computing devices that is a member of the second set and not the first set, instructing the grid virtual machine instance running on the computing device to process one of the plurality of subtasks.

3. The method of claim 1, wherein the customer load factor is computed by examining metrics selected from the group consisting of:

processor usage, network usage, storage usage, Unix load average, RAM usage, and ping time.

4. The method of claim 1, further comprising:

computing the composite load factor.

5. The method of claim 1, wherein the work request is from an external client.

6. The method of claim 5, wherein the external client is a hypervisor included in the cloud computing system.

7. The method of claim 1, wherein the work request is encoded in a network protocol.

8. The method of claim 7, wherein the network protocol is selected from the group consisting of:

JSON, HTTP, SOAP, XML, and SUN RPC.

9. The method of claim 1, wherein the work request is specified in the form of instructions to be run by the grid virtual machine.

10. The method of claim 9, wherein the instructions are written in a computer programming language.

11. The method of claim 10, wherein the computer programming language is selected from the group consisting of:

C, C++, C#, Java, Python, Perl, Javascript, and Visual Basic.

12. The method of claim 1, wherein the work request is encoded as byte codes that are read and interpreted by the grid virtual machine.

13. The method of claim 1, wherein dividing the work request includes examining contents of the work request and identifying portions of the work request to be performed in parallel.

14. The method of claim 1, further comprising:

examining the customer load factor of each of the plurality of computing devices to produce a second set of computing devices with customer load factors below a second threshold different from the first threshold;

for each of the plurality of computing devices that is a member of the first set and not the second set, instructing the grid virtual machine instance running on the computing device to stop processing the subtask it was previously instructed to process; and

for each of the plurality of computing devices that is a member of the second set and not the first set, instructing the grid virtual machine instance running on the computing device to process one of the plurality of subtasks.

15. A cloud computing system for utilizing spare resources, the system comprising:

one or more processors;

a computing device configured to run virtual machine instances, wherein the computing device includes a hypervisor program for managing the virtual machine instances, and wherein the computing device is executable on the one or more processors; and

a grid virtual machine instance run by the hypervisor program on the computing device,

wherein the computing device receives a work request via a first application programming interface (API), determines a composite load factor on each of the plurality of computing devices, examines the composite load factor on each of the plurality of computing devices to produce a first set of computing devices with composite load factors below a threshold, divides the work request into a plurality of subtasks, and instructs, via a second API, the grid virtual machine instance running on each computing device in the first set of computing devices to process one of the plurality of subtasks, and wherein the composite load factor represents the overall load on the respective computing device.

16. The cloud computing system of claim 15, wherein the computing device examines the customer load factor of each of the plurality of computing devices to produce a second set of computing devices with customer load factors below the threshold, wherein for each of the plurality of computing devices that is a member of the first set and not the second set, the computing device instructs the grid virtual machine instance running on the computing device to stop processing the subtask it was previously instructed to process, and wherein for each of the plurality of computing devices that is a member of the second set and not the first set, the computing device instructs the grid virtual machine instance running on the computing device to process one of the plurality of subtasks.

17. The cloud computing system of claim 15, wherein the computing device examines the customer load factor of each of the plurality of computing devices to produce a second set of computing devices with customer load factors below a second threshold different from the first threshold, wherein for each of the plurality of computing devices that is a member of the first set and not the second set, the computing device instructs the grid virtual machine instance running on the computing device to stop processing the subtask it was previously instructed to process, and wherein for each of the plurality of computing devices that is a member of the second set and not the first set, the computing device instructs the grid virtual machine instance running on the computing device to process one of the plurality of subtasks.

18. The cloud computing system of claim 15, wherein the computing device computes the customer load factor by examining metrics selected from the group consisting of:

processor usage, network usage, storage usage, Unix load average, RAM usage, and ping time.

19. The cloud computing system of claim 15, wherein the work request is from an external client.

20. The cloud computing system of claim 15, wherein the work request is encoded in a network protocol.