Rolling capacity upgrade control

Patent metadata

Field	Value
Publication number	US-2016197844-A1
Application number	US-201514588653-A
Country	US
Kind code	A1
Filing date	Jan 2, 2015
Priority date	Jan 2, 2015
Publication date	Jul 7, 2016
Grant date	—

Abstract

In order to upgrade a source farm to a target farm, the target farm is created with an initial increment of resource capacity that is a fraction of a final resource capacity of the target farm. A corresponding increment of the databases attached to the source farm are then switched from the source farm to the target farm. The resource capacity at the source farm, that was used to service the moved databases, is then removed from the source farm and added to the target farm, and another increment of the databases are switched from the source farm to the target farm. This continues until a full capacity of resources is deployed at the target farm, and the databases are fully moved from the source farm to the target farm.

First claim

What is claimed is:

1 . A computing system for controlling deployment of computing resources, comprising:
an upgrade controller sensing a resource deployment input indicative of a request to deploy computing resources in a target server farm, and accessing an increment indicator indicative of a fixed increment of the computing resources, that is less than a final resource capacity, to be deployed in the target server farm; a resource capacity reassignment component removing, in response to a call from the upgrade controller, the fixed increment of computing resources from a source server farm and deploying the fixed increment of computing resources to the target server farm; and a database move component that determines whether the target server farm has deployed adequate resource capacity to service a fixed increment of databases, and, if so, moves the fixed increment of databases from the source server farm to the target server farm, and, if not, waits for the resource capacity reassignment component to reassign additional computing resources from the source server farm to the target server farm.

2 . The computing system of claim 1 wherein the computing resources comprise at least one of hardware computing systems and virtual machines implemented by the hardware computing systems.

3 . The computing system of claim 2 and further comprising:
a goal processing component that adjusts a number of virtual machines on a given server farm based on a resource capacity goal for the given server farm, and wherein the resource capacity reassignment component removes the fixed increment of computing resources from the source server farm by reducing the resource capacity goal, for the source server farm, by the fixed increment.

4 . The computing system of claim 3 wherein the resource capacity reassignment component deploys the fixed increment of computing resources at the target server farm by increasing the resource capacity goal, for the target server farm, by the fixed increment.

5 . The computing system of claim 2 wherein the computing resources implement a multi-tenant service on the source server farm and an upgraded multi-tenant service on the target server farm.

6 . The computing system of claim 5 and further comprising:
a set of request routers that route service requests for tenants in the source server farm to computing resources in the source server farm, and wherein the upgrade controller identifies, as the fixed increment, a fractional part of the set of request routers.

7 . The computing system of claim 5 wherein the upgrade controller identifies the fixed increment, independently of a workload level on the hardware computing devices and virtual machines on the source server farm.

8 . The computing system of claim 5 wherein the upgrade controller identifies the fixed increment based on a number of the databases operably coupled to the source server farm.

9 . The computing system of claim 5 wherein the upgrade controller identifies the fixed increment as a fixed percent of a final resource capacity of the target server farm.

10 . The computing system of claim 5 and further comprising:
a farm creation component that creates the target server farm, with no computing resources assigned, in response to a call from the upgrade controller.

11 . The computing system of claim 10 and further comprising:
a farm tear down component that removes a final increment of the computing resources from the source server farm, after a final increment of the databases has been removed from the source server farm.

12 . The computing system of claim 1 wherein the source server farm and the target server farm both implement multi-tenant, cloud-based services.

13 . A method, comprising:
sensing a reassignment input to reassign all computing resources from a source server farm to a target server farm; incrementally reassigning all of the computing resources from the source server farm to the target server farm, in a plurality of fixed increments; and after reassigning each fixed increment of the computing resources to the target server farm, moving a set of databases, corresponding to a previously reassigned fixed increment of computing resources, from the source server farm to the target server farm, for servicing at the target server farm.

14 . The method of claim 13 wherein incrementally reassigning comprises:
identifying the fixed increment;
removing the fixed increment of computing resources from the source server farm; and
adding the fixed increment of computing resources, removed from the source server farm, to the target server farm.

15 . The method of claim 14 wherein the source and target server farms vary a level of virtual machine computing resources generated based on a capacity goal, and wherein removing the fixed increment of computing resources from the source server farm comprises lowering the capacity goal for the source server farm by the fixed increment.

16 . The method of claim 15 wherein adding the fixed increment of computing resources to the target server farm comprises increasing the capacity goal for the target server farm by the fixed increment.

17 . The method of claim 14 and further comprising:
after all databases are moved from the source server farm to the target server farm, removing all remaining computing resources from the source server farm.

18 . The method of claim 13 wherein incrementally reassigning comprises:
incrementally reassigning all of the computing resources from the source server farm to the target server farm, in the plurality of fixed increments, independently of a processing load on the source server farm.

19 . A computer readable storage medium that stores computer executable instructions which, when executed by the computer, cause the computer to perform a method, comprising:
sensing a reassignment input to reassign all computing resources from a source server farm to a target server farm; incrementally reassigning all of the computing resources from the source server farm to the target server farm, in a plurality of fixed increments by removing the fixed increment of computing resources from the source server farm and adding the fixed increment of computing resources, removed from the source server farm, to the target server farm; and after reassigning each fixed increment of the computing resources to the target server farm, moving a set of databases, corresponding to an immediately previously reassigned fixed increment of computing resources, from the source server farm to the target server farm, for servicing at the target server farm.

20 . The computer readable storage medium of claim 19 wherein the source and target server farms vary a level of virtual machine computing resources generated based on a capacity goal, and wherein removing the fixed increment of computing resources from the source server farm comprises lowering the capacity goal for the source server farm by the fixed increment, and wherein adding the fixed increment of computing resources to the target server farm comprises increasing the capacity goal for the target server farm by the fixed increment.

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Smith Bradley S
• Kona Kalyan K
• Paranthaman Sundaravadivelan
• Jauhari Ankur

Classifications

• H04L41/0896

  Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities (flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76) · CPC title

• G06F9/5061

  Partitioning or combining of resources · CPC title

• H04L41/5096

  wherein the managed service relates to distributed or central networked applications · CPC title

• H04L67/1004 Primary

  Server selection for load balancing · CPC title

• G06F9/45558

  Hypervisor-specific management and integration aspects · CPC title