Defragmenting clusters with reserved resources

• US 9,350,800 B2
• Filed: 06/05/2013
• Issued: 05/24/2016
• Est. Priority Date: 06/05/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Lang, Willis, Teletia, Nikhil, Kalhan, Ajay, Halverson, Alan Dale, Varanasi, Prashant
• Primary Examiner(s): SHINGLES, KRISTIE D
• Application Number:
  US13/910,312
• Publication Number:
  US 20140365658A1
• Time in Patent Office: 1,084 Days
• Field of Search: 709/223, 709/226, 709/201
• US Class Current: 1/1
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests
• CPC Class Codes: H04L 67/1076 Resource dissemination mechanisms or network resource keeping policies for optimal resource availability in the overlay network
• CPC Class Codes: H04L 67/108 characterised by resources being split in blocks or fragments
• CPC Class Codes: H04L 67/34 involving the movement of software or configuration parameters network booting or remote initial program loading [RIPL] G06F9/4416
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

Defragmenting a cluster service to service additional capacity requests on the service. A method includes determining an amount of server resources needed for an additional deployment reservation request for a new deployment or increasing reservation of resources of an existing deployment. The method further includes determining a server that currently does not have capacity to service the additional deployment reservation request. The method further includes determining how resources on the server can be freed up by moving other replicas of other deployments on the server to other servers to allow the server to service the additional deployment reservation request.

Claims

1. In a distributed computing environment, a method of defragmenting a cluster service to service additional capacity requests on the cluster service, the method comprising:

determining an amount of server resources needed for an additional deployment reservation request for a new deployment or increasing reservation of resources of an existing deployment;

determining a server that currently does not have capacity to service the additional deployment reservation request;

determining how resources on the server can be freed up by moving one or more replicas of other deployments on the server to other servers to allow the server to service the additional deployment reservation request; and

moving at least one replica of the one or more replicas of the other deployments on the server to at least one of the other servers, based on determining the server fails to have capacity to service the additional deployment reservation request, and to allow the server to service the additional deployment reservation request, wherein the at least one replica is moved to the at least one of the other servers that is determined to be on at least one of a different upgrade domain or a different fault domain than another replica of a same deployment as the at least one replica.

2. The method of claim 1, wherein the additional deployment reservation request is an upgrade request to reserve additional resources for an existing deployment.

3. The method of claim 1, wherein the additional deployment reservation request is a new deployment request requesting a reservation of resources for a new deployment.

4. The method of claim 1, wherein the at least one replica is moved to the at least one of the other servers that is determined to be on the different upgrade domain than said another replica of the same deployment as the at least one replica.

5. The method of claim 1, wherein the at least one replica is moved to the at least one of the other servers that is determined to be on the different fault domain than said another replica of the same deployment as the at least one replica.

6. The method of claim 1, wherein determining the server that currently does not have capacity to service the additional capacity deployment request comprises ranking the other servers in a cluster by free capacity and selecting at least one of the other servers with the most free capacity.

7. The method of claim 1 further comprising, determining the at least one of the other servers to which replicas from the other deployments can be moved by ranking the one or more servers by available capacity and determining to move the at least one replica to the at least one of the other servers which is determined to have a least available capacity that could fit the replica.

8. The method of claim 1, further comprising, determining to move the at least one replica based at least on a cost of moving the at least one replica as compared to the cost for moving at least one other replica.

9. The method of claim 8, wherein the cost is determined by sizes of replicas.

10. The method of claim 8, wherein the cost is determined by at least a number of replicas to move.

11. The method of claim 1, further comprising removing the server from further replica movement consideration based on already having determined to move resources from the server.

12. The method of claim 1, further comprising, taking a snapshot of a cluster environment, and wherein the method is performed using the snapshot.

13. In a distributed computing environment, a system for defragmenting a cluster service to service additional capacity requests on the service the system comprising:

one or more processors; and

one or more computer readable media, wherein the one or more computer readable media comprise computer executable instructions that when executed by at least one of the one or more processors cause at least one of the one or more processors to perform the following;

determining an amount of server resources needed for an additional deployment reservation request for a new deployment or increasing reservation of resources of an existing deployment;

determining a server that currently does not have capacity to service the additional deployment reservation request;

determining how resources on the server can be freed up by moving one or more replicas of other deployments on the server to other servers to allow the server to service the additional deployment reservation request; and

moving at least one replica of the one or more replicas of the other deployments on the server to at least one of the other servers, based on determining the server fails to have capacity to service the additional deployment reservation request, and to allow the server to service the additional deployment reservation request, wherein the at least one replica is moved to the at least one of the other servers that is determined to be on at least one of a different upgrade domain or a different fault domain than another replica of a same deployment as the at least one replica.

14. The system of claim 13, wherein the at least one replica is moved to the at least one of the other servers that is determined to be on a different upgrade domain than said another replica of the same deployment as the at least one replica.

15. The system of claim 13, wherein determining the server that currently does not have capacity to service the additional capacity deployment request comprises ranking servers in a cluster by free capacity and selecting the server with the most free capacity.

16. The system of claim 13, further comprising, determining the at least one server to which the at least one replica can be moved by ranking the other servers by available capacity and determining to move the at least one replica to the at least one of the other servers determined to have a least capacity that could fit the replica.

17. The system of claim 13, further comprising, determining the at least one replica to move from the server by a cost of moving the at least one replica as compared to a cost for moving other replicas.

18. The system of claim 13, further comprising removing the server from further replica movement consideration based on already having determined to move resources from the server.

19. The system of claim 13, further comprising, taking a snapshot of a cluster environment, and wherein the steps of claim 13 are performed using the snapshot.

20. One or more hardware storage device having stored computer executable instructions that are executable by one or more processors of a computing system to cause the computing system to perform a method of defragmenting a cluster service to service additional capacity requests on the cluster service by performing the following:

determining an amount of server resources needed for an additional deployment reservation request for a new deployment or increasing reservation of resources of an existing deployment;

determining a server that currently does not have capacity to service the additional deployment reservation request;

determining how resources on the server can be freed up by moving one or more replicas of other deployments on the server to other servers to allow the server to service the additional deployment reservation request; and

moving at least one replica of the one or more replicas of the other deployments on the server to at least one of the other servers, based on determining the server fails to have capacity to service the additional deployment reservation request, and to allow the server to service the additional deployment reservation request, wherein the at least one replica is moved to the at least one of the other servers that is determined to be on at least one of a different upgrade domain or a different fault domain than another replica of a same deployment as the at least one replica.