Instance level server application monitoring, load balancing, and resource allocation

• US 9,323,628 B2
• Filed: 10/08/2013
• Issued: 04/26/2016
• Est. Priority Date: 10/09/2012
• Status: Active Grant

• Current Assignee: DH2I Company
• Original Assignee: DH2I Company
• Inventors: Ngo, Thanh Quy, Revitch, Samuel
• Primary Examiner(s): Truong, Loan L. T.
• Application Number:
  US14/048,241
• Publication Number:
  US 20140149784A1
• Time in Patent Office: 931 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/2002 where interconnections or communication control functionality are redundant flexible arrangements for bus networks involving redundancy H04L12/40176
• CPC Class Codes: H04L 41/0668 by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure
• CPC Class Codes: H04L 41/0836 to enhance reliability, e.g. reduce downtime
• CPC Class Codes: H04L 41/5025 by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30

Abstract

A system and methodology to monitor system resources for a cluster computer environment and/or an application instance allows user to defined failover policies that take appropriate corrective actions when a predefined threshold is met. An engine comprising failover policies and mechanisms to define resource monitoring, consumption, allocation, and one or more thresholds for a computer server environment to identify capable servers and thereafter automatically transition an application between multiple servers so as to ensure the application is continually operating within the defined metrics.

Claims

1. A system for application instantiation failover optimization, comprising:

a plurality of servers configured into a server cluster, wherein the plurality of servers is comprised of a combination of one or more physical machines and one or more virtual machines, and wherein each server within the server cluster includes a local monitor engine, a cluster monitor engine, a performance monitor engine, and an alert handler engine;

at least one instantiation of an application configured to operate on an instance container on at least one of the plurality of servers and wherein the instance container is a virtual host that operates on top of, and move freely between, the one or more virtual machines and the one or more physical servers of the server cluster and wherein the instance container instantiates at least one virtual internet protocol address distinct from an internet protocol address of the virtual machines on which it operates;

a service level agreement associated with the at least one instantiation of the application identifying minimum application performance requirements for the instantiation of the application; and

a failover policy identifying by the cluster monitor engine one or more servers within the server cluster operable to accept the virtual host and instantiation of the application and meet the service level agreement and wherein responsive to the instantiation of the application failing to meet the service level agreement, failing over the virtual host and instantiation of the application to one of the identified one or more servers within the server cluster.

2. The system for application instantiation failover optimization according to claim 1, wherein the service level agreement includes server processing characteristics, memory allocation, and network activity.

3. The system for application instantiation failover optimization according to claim 1, wherein the one or more servers identified in the failover policy are examined to determine whether upon instantiation of the application, they can meet the service level agreement.

4. The system for application instantiation failover optimization according to claim 3, wherein the one or more servers identified in the failover policy are sequentially examined.

5. The system for application instantiation failover optimization according to claim 3, wherein responsive to determining at least one of the servers identified in the failover policy can meet the service level agreement upon instantiation of the application, initiating the failover of the instantiation of the application.

6. The system for application instantiation failover optimization according to claim 3, wherein responsive to determining that none of the servers identified in the failover policy can meet the service level agreement upon instantiation of the application, maintaining the instantiation of the application on the at least one of the plurality of servers.

7. The system for application instantiation failover optimization according to claim 1, wherein responsive to determining that the instantiation of the application is failing to meet the service level agreement identifying by the local monitoring engine other server resources that can be modified to enable service level agreement compliance.

8. A method for application instance failover optimization, comprising:

monitoring, in a server cluster wherein the server cluster is comprised of a combination of one or more physical machines and one or more virtual machines, application instance performance of an application instance operating on an instance container on a first node within the server cluster wherein the application instance and the instance container can operate on top of, and move freely between, the one or more virtual machines and the one or more physical servers of the server cluster, and wherein the instance container instantiates at least one virtual internet protocol address distinct from an internet protocol address of the one or more virtual machines on which it operates; and

responsive to identifying application instance performance degradation on the first node within the server cluster, determining application instance resource availability within other nodes within the server cluster, and, when with resource availability confirmed, initiating failover of the instance container and application instance.

9. The method for application instance failover optimization according to claim 8, wherein application instance performance is measured against a predetermined service level agreement.

10. The method for application instance failover optimization according to claim 9, wherein application instance resource availability is based on the predetermined service level agreement.

11. The method for application instance failover optimization according to claim 8, further comprising managing by a cluster monitor server application tasking within the server cluster.

12. The method for application instance failover optimization according to claim 8, wherein responsive to confirming resource availability on a second node within the server cluster capable of supporting the application instance, initiating failover to the second node.

13. The method for application instance failover optimization according to claim 8, wherein other nodes within the server cluster are identified by a predetermined failover policy.

14. The method for application instance failover optimization according to claim 8, wherein responsive to being unable to identify another node with instance resource available, maintaining the application instance performance degradation on the first node.

15. The method for application instance failover optimization according to claim 8, further comprising identifying resource availability on the first node based on an instance failover policy.

16. The method for application instance failover optimization according to claim 15, further comprising reallocating resources on the first node based on the instance failover policy.

17. A system for application instance failover resource determination, comprising:

a plurality of servers forming a server cluster wherein the plurality of servers is comprised of a combination of one or more physical machines and one or more virtual machines, and wherein at least one server within the server cluster includes an application instance operating on top of a virtual host wherein the virtual host can operate on top of, and move freely between the one or more virtual machines and the one or more physical servers of the server cluster, and wherein the instance container instantiates at least one virtual internet protocol address distinct from an internet protocol address of the virtual machines on which it operates;

a performance monitor engine operable on each of the plurality of servers to monitor application instance performance based on a service level agreement; and

a cluster monitor engine operable to identify other servers within the server cluster having sufficient resources available to meet application instance performance requirements in compliance with the service level agreement, and wherein upon identifying degraded application instance performance, initiating a failover of the virtual host and application instance to at least one of the other servers in the server cluster sufficient to meet application instance performance in compliance with the service level agreement.

18. The system for application instance failover resource determination according to claim 17, further comprising a local monitor engine operable to determine on the server within the server cluster local instance resource allocation and failover priority.

19. The system for application instance failover resource determination according to claim 18, wherein responsive to identifying degraded instance performance reallocating server resources based an application failover priority policy.

20. The system for application instance failover resource determination according to claim 17, wherein upon failing to identify other servers having sufficient resources available to meet application performance requirements in compliance with the service level agreement, maintain degraded application instance performance.