System and method for dynamic scaling based on a reference architecture

• US 9,116,753 B2
• Filed: 07/08/2012
• Issued: 08/25/2015
• Est. Priority Date: 07/08/2012
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Devi, Yogesh
• Primary Examiner(s): An, Meng
• Assistant Examiner(s): Zhao, Bing
• Application Number:
  US13/543,856
• Publication Number:
  US 20140013323A1
• Time in Patent Office: 1,143 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 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system

Abstract

A method includes, on a virtualization platform, creating a first runtime instance of a virtual appliance. The virtual appliance includes a solution stack of software components selected from a reference architecture. The method further includes, responsive to satisfaction of at least one performance criterion, dynamically scaling-out the solution stack to one or more additional runtime instances of the virtual appliance. The dynamically scaling-out includes distributing a subset of the solution stack to each of the first runtime instance and the one or more additional runtime instances.

Claims

1. A method comprising:

on a virtualization platform, creating a first runtime instance of a virtual appliance;

wherein the virtual appliance comprises an image file that includes a pre-configured operating-system environment packaged with a solution stack of software components;

wherein the solution stack of software components are selected software components of a reference architecture comprising a plurality of software components;

responsive to satisfaction of at least one performance criterion, dynamically scaling-out the solution stack to one or more additional runtime instances of the virtual appliance such that, in total, more physical resources are provisioned to the solution stack; and

wherein the dynamically scaling-out comprises distributing, based on scaling rules, a subset of the solution stack to each of the first runtime instance and the one or more additional runtime instances;

wherein the scaling rules define a plurality of explosion levels;

wherein, for each explosion level in the plurality of explosion levels, the scaling rules specify a total number of runtime instances, a particular subset of the solution stack that should be distributed to each runtime instance, and provisioning requirements for each runtime instance;

wherein, for each runtime instance of the first runtime instance and the one or more additional runtime instances, the distributing comprises;

activating software components of the solution stack distributed to the runtime instance; and

deactivating software components of the solution stack not distributed to the runtime instance.

2. The method of claim 1, wherein the dynamically scaling-out comprises creating the one or more additional runtime instances based on the scaling rules.

3. The method of claim 2, wherein the dynamically scaling-out comprises provisioning physical resources to the one or more runtime instances based on the scaling rules.

4. The method of claim 2, wherein:

wherein the plurality of explosion levels comprise discrete levels of progressively increasing throughput that are defined in terms of runtime instances; and

the dynamically scaling-out comprises transitioning from one level of the plurality of explosion levels to a higher level of the plurality of explosion levels.

5. The method of claim 1, comprising dynamically scaling-in the solution stack to the first runtime instance.

6. The method of claim 1, wherein the reference architecture comprises a service-oriented architecture (SOA).

7. The method of claim 1, wherein the at least one performance criterion is selected from the group consisting of processor utilization and memory utilization.

8. The method of claim 1, comprising facilitating communication among the software components.

9. The method of claim 8, wherein the facilitating comprises utilizing distribution proxies associated with each of the software components.

10. An information handling system comprising:

a computer system comprising physical resources; and

a hypervisor resident on the computer system, wherein the hypervisor is operable to;

create a first runtime instance of a virtual appliance;

wherein the virtual appliance comprises an image file that includes a pre-configured operating-system environment packaged with a solution stack of software components;

wherein the solution stack of software components are selected software components of a reference architecture comprising a plurality of software components;

responsive to satisfaction of at least one performance criterion, dynamically scale-out the solution stack to one or more additional runtime instances of the virtual appliance such that, in total, more physical resources are provisioned to the solution stack;

wherein the dynamic scale-out comprises distribution, based on scaling rules, of a subset of the solution stack to each of the first runtime instance and the one or more additional runtime instances;

wherein the scaling rules define a plurality of explosion levels;

wherein, for each explosion level in the plurality of explosion levels, the scaling rules specify a total number of runtime instances, a particular subset of the solution stack that should be distributed to each runtime instance, and provisioning requirements for each runtime instance;

wherein, for each runtime instance of the first runtime instance and the one or more additional runtime instances, the distribution comprises;

activation of software components of the solution stack distributed to the runtime instance; and

deactivation of software components of the solution stack not distributed to the runtime instance.

11. The information handling system of claim 10, wherein the virtual appliance comprises a metadata store, the metadata store comprising the scaling rules.

12. The information handling system of claim 11, wherein the dynamic scale-out comprises creation of the one or more additional runtime instances based on the scaling rules.

13. The information handling system of claim 11, wherein the dynamic scale-out comprises provision of at least a portion of the physical resources to the one or more runtime instances based on the scaling rules.

14. The information handling system of claim 11, wherein:

wherein the plurality of explosion levels comprise discrete levels of progressively increasing throughput that are defined in terms of a number of runtime instances; and

the dynamic scale-out comprises a transition from one level of the plurality of explosion levels to a higher level of the plurality of explosion levels.

15. The information handling system of claim 10, wherein the virtual appliance comprises an instance map that maps each software component of the solution stack to a runtime instance selected from among the first runtime instance and the one or more additional runtime instances.

16. The information handling system of claim 10, wherein the hypervisor is operable to scale in the solution stack to the first runtime instance.

17. The information handling system of claim 10, wherein the reference architecture comprises a service-oriented architecture (SOA).

18. The information handling system of claim 10, wherein the at least one performance criterion is selected from the group consisting of processor utilization and memory utilization.

19. The information handling system of claim 10, wherein the hypervisor is operable to facilitate communication among the software components via utilization of proxies associated with each of the software components.

20. A computer-program product comprising a non-transitory computer-usable medium having computer-readable program code embodied therein, the computer-readable program code adapted to be executed to implement a method comprising:

creating a first runtime instance of a virtual appliance;

wherein the virtual appliance comprises an image file that includes a pre-configured operating-system environment packaged with a solution stack of software components;

wherein the solution stack of software components are selected software components of a reference architecture comprising a plurality of software components;

responsive to satisfaction of at least one performance criterion, dynamically scaling-out the solution stack to one or more additional runtime instances of the virtual appliance such that, in total, more physical resources are provisioned to the solution stack; and

wherein the dynamically scaling-out comprises distributing, based on scaling rules, a subset of the solution stack to each of the first runtime instance and the one or more additional runtime instances;

wherein the scaling rules define a plurality of explosion levels;

wherein, for each explosion level in the plurality of explosion levels, the scaling rules specify a total number of runtime instances, a particular subset of the solution stack that should be distributed to each runtime instance, and provisioning requirements for each runtime instance;

wherein, for each runtime instance of the first runtime instance and the one or more additional runtime instances, the distributing comprises;

activating software components of the solution stack distributed to the runtime instance; and

deactivating software components of the solution stack not distributed to the runtime instance.