Management of demand for virtual computing resources

Patent metadata

Field	Value
Publication number	US-2018246744-A1
Application number	US-201815900712-A
Country	US
Kind code	A1
Filing date	Feb 20, 2018
Priority date	Nov 24, 2014
Publication date	Aug 30, 2018
Grant date	—

Abstract

Systems and methods for managing demand for virtual computing resources are disclosed. A demand forecasting service can obtain and analyze historical demand data for purposes of predicting future demand. The analysis includes identifying untargeted demand corresponding to requests for virtual machine instances that can be fulfilled by any availability zone of a set of zones. The demand forecasting service may provide predictions of future demand including information regarding future untargeted demand, thereby enabling efficient allocation of computing resources among various availability zones to meet the future demand.

First claim

What is claimed is:

1 . A computer-implemented method comprising:
receiving one or more requests by a web-based service for a first set of one or more virtual machine resources; allocating a second set of one or more virtual machine resources within an availability zone in response to the one or more requests based, at least in part, on a past number of virtual machine resources allocated within the availability zone to service requests received by the web-based service.

2 . The computer-implemented method of claim 1 , wherein the one or more requests are untargeted requests.

3 . The computer-implemented method of claim 1 , wherein the availability zone comprises a plurality of data centers.

4 . The computer-implemented method of claim 3 , wherein the one or more requests target a specified data center within the plurality of data centers.

5 . The computer-implemented method of claim 4 , wherein allocating the second set of one or more virtual machine resources comprises determining, for individual data centers within the plurality of data centers, a portion of the second set of one or more virtual machine resources to allocate to the individual data center.

6 . The computer-implemented method of claim 1 further comprising determining, based at least in part on the one or more requests for the first set of one or more virtual machine resources, a ratio between untargeted requests serviced by the availability zone and targeted requests serviced by the availability zone.

7 . A system comprising:
a data store configured to store computer-executable instructions; and a processor in communication with the data store, the processor configured to execute the computer-executable instructions to at least:
receive a plurality of requests for virtual machine resources;
determine, based at least in part on the plurality of requests, future demand for virtual machine resources within an availability zone; and
allocate additional virtual machine resources within the availability zone based at least in part on the future demand.

8 . The system of claim 7 , wherein the plurality of requests comprises one or more untargeted requests, the one or more untargeted requests indicating that any data center within the availability zone can service the request.

9 . The system of claim 7 , wherein the future demand comprises future targeted demand and future untargeted demand.

10 . The system of claim 7 , wherein the availability zone corresponds to a distinct geographic region.

11 . The system of claim 7 , wherein the processor is further configured to at least obtain data regarding a free capacity of one or more data centers within the availability zone.

12 . The system of claim 11 , wherein the processor is configured to allocate the additional virtual machine resources within the availability zone based at least in part on the free capacity of one or more data centers within the availability zone.

13 . The system of claim 7 , wherein the processor is configured to determine the future demand based at least in part on a forecasting model.

14 . The system of claim 7 , wherein the plurality of requests comprises historical requests for virtual machine resources within the availability zone.

15 . A non-transitory computer-readable medium storing computer-executable instructions that, when executed by a processor, perform operations comprising:
predicting, based at least in part on historical demand associated with an availability zone, future demand associated with the availability zone; and allocating one or more additional virtual machine resources within the availability zone based at least in part on the future demand.

16 . The non-transitory computer-readable medium of claim 15 , the operations further comprising identifying, based at least in part on historical demand associated with a plurality of availability zones, the historical demand associated with the availability zone.

17 . The non-transitory computer-readable medium of claim 15 , wherein the historical demand associated with the availability zone comprises targeted demand for virtual machine resources within the availability zone and untargeted demand for virtual machine resources within the availability zone.

18 . The non-transitory computer-readable medium of claim 15 , wherein the future demand associated with the availability zone comprises future targeted demand and future untargeted demand.

19 . The non-transitory computer-readable medium of claim 15 , wherein the availability zone is a first availability zone, the operations further comprising:
identifying historical demand associated with a second availability zone; predicting, based at least in part on the historical demand associated with the second availability zone, future demand associated with the second availability zone; and allocating the one or more additional virtual machine resources between the first availability zone and the second availability zone based at least in part on the predicted future demands.

20 . The non-transitory computer-readable medium of claim 15 , wherein the availability zone is associated with one or more data centers within a distinct geographic region.

Assignees

• Amazon Tech Inc

Inventors

• Li Xiao Yu
• Cary Charles Ian Ormsby

Classifications

• G06F2009/45591

  Monitoring or debugging support · CPC title

• G06F2009/45562

  Creating, deleting, cloning virtual machine instances · CPC title

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title

• G06F9/45533 Primary

  Hypervisors; Virtual machine monitors · CPC title

• G06F2009/45575

  Starting, stopping, suspending or resuming virtual machine instances · CPC title