Predicting resource demand in computing environments

Patent metadata

Field	Value
Publication number	US-10887176-B2
Application number	US-201715473729-A
Country	US
Kind code	B2
Filing date	Mar 30, 2017
Priority date	Mar 30, 2017
Publication date	Jan 5, 2021
Grant date	Jan 5, 2021

Abstract

In some examples, a method can involve collecting resource consumption data for resource consumer objects associated with hosts in a computing environment. The method can involve identifying, for each respective host, a respective set of resource consumer objects at the host and, based on the resource consumption data, determining a projected resource consumption history for each host, the projected resource consumption history being based on a combined resource consumption, over a period of time, associated with the respective set of resource consumer objects currently hosted a the host. The method can involve calculating a projected resource availability for each host based on a respective resource capability of the host and the projected resource consumption history for the host, and selecting a particular host for a resource consumer object based on the projected resource availability of each host.

First claim

What is claimed is:

1. A method to manage resources in a computing environment, the method comprising:
collecting respective resource consumption data for each of a plurality of resource consumer objects associated with a plurality of hosts in the computing environment;
for each respective host from the plurality of hosts, identifying a respective set of resource consumer objects currently hosted by the respective host;
based on the respective resource consumption data, determining a projected resource consumption history for each respective host, wherein the projected resource consumption history:
comprises a combined resource consumption, over a predetermined period of time, associated with the respective set of resource consumer objects that are currently being hosted by the respective host, the predetermined period of time immediately preceding a current time; and
is a historical resource consumption reconstructed based on the respective host hosting the respective set of the currently hosted resource consumer objects through the predetermined period of time in entirety wherein the respective host did not host one or more resource consumer objects of the respective set of currently hosted resource consumer objects through the predetermined period of time in entirety;

calculating a projected resource availability for each respective host based on a respective resource capability of the respective host and the projected resource consumption history for the respective host; and
selecting, from the plurality of hosts, a particular host for a resource consumer object based on the projected resource availability calculated for each respective host.

2. The method of claim 1 , further comprising:
deploying the resource consumer object at the particular host; and
allocating, to the resource consumer object, one or more resources associated with the particular host.

3. The method of claim 2 , wherein the computing environment comprises a pool of resources, and wherein each respective host from the plurality of hosts is allocated a respective set of resources from the pool of resources.

4. The method of claim 3 , wherein the respective host allocates at least a portion of the respective set of resources to the respective set of resource consumer objects currently hosted by the respective host, and wherein the one or more resources allocated to the resource consumer object comprise at least a portion of the respective set of resources allocated to the particular host.

5. The method of claim 4 , wherein the pool of resources comprises at least one of storage resources, compute resources, virtualization resources, backup resources, memory, bandwidth resources, distributed resource clusters, and data switching or routing resources.

6. The method of claim 4 , wherein the plurality of resource consumer objects comprises at least one of virtual machines and software containers, and wherein the plurality of hosts comprises at least one of hypervisors, servers, datacenters, and infrastructure appliances.

7. The method of claim 1 , wherein the respective resource consumption data is collected at one or more time intervals over the predetermined period of time.

8. The method of claim 1 , further comprising:
based on the respective resource consumption data, determining a respective resource consumption trend for each of the plurality of resource consumer objects corresponding to the predetermined period of time;
determining a current resource consumption for each of the plurality of resource consumer objects; and
calculating a projected resource demand for each of the plurality of resource consumer objects based on at least one of the respective resource consumption trend and the current resource consumption.

9. The method of claim 1 , wherein determining the projected resource consumption history for each respective host is based on the projected resource demand for each of the plurality of resource consumer objects, and wherein the resource consumption associated with the particular resource consumer object is determined based on at least one of the respective resource consumption data and a current resource consumption measurement associated with the particular resource consumer object.

10. A system to manage resources in a computing environment, the system comprising:
one or more processors; and
at least one computer-readable storage medium having stored therein instructions which, when executed by the one or more processors, cause the system to:
collect respective resource consumption data for each of a plurality of resource consumer objects associated with a plurality of hosts in the computing environment;

for each respective host from the plurality of hosts, identify a respective set of resource consumer objects currently hosted by the respective host;
based on the respective resource consumption data, determining a projected resource consumption history for each respective host, wherein the projected resource consumption history:
comprises a combined resource consumption, over a predetermined period of time, associated with the respective set of resource consumer objects that are currently being hosted by the respective host, the predetermined period of time immediately preceding a current time; and
is a historical resource consumption reconstructed based on the respective host hosting the respective set of the currently hosted resource consumer objects through the predetermined period of time in entirety wherein the respective host did not host one or more resource consumer objects of the respective set of currently hosted resource consumer objects through the predetermined period of time in entirety;

calculate a projected resource availability for each respective host based on a respective resource capability of the respective host and the projected resource consumption history for the respective host; and
select, from the plurality of hosts, a particular host for a resource consumer object based on the projected resource availability calculated for each respective host.

11. The system of claim 10 , the at least one computer-readable storage medium storing additional instructions which, when executed by the one or more processors, cause the system to:
deploy the resource consumer object at the particular host; and
allocate, to the resource consumer object, one or more resources associated with the particular host.

12. The system of claim 11 , wherein the computing environment comprises a pool of resources, and wherein each respective host from the plurality of hosts is allocated a respective set of resources from the pool of resources.

13. The system of claim 12 , wherein the one or more resources comprise at least a portion of the respective set of resources allocated to the particular host.

14. The system of claim 10 , the at least one computer-readable storage medium storing additional instructions which, when executed by the one or more processors, cause the system to:
based on the respective resource consumption data, determine a respective resource consumption trend for each of the plurality of resource consumer objects corresponding to the predetermined period of time;
determine a current resource consumption for each of the plurality of resource consumer objects; and
calculate a projected resource demand for each of the plurality of resource consumer objects based on at least one of the respective resource consumption trend and the current resource consumption.

15. A non-transitory computer-readable storage medium to manage resources in a computing environment, the non-transitory compute

Assignees

• Hewlett Packard Entpr Dev Lp

Inventors

• Fitzgerald Marty
• Schindler Jiri

Classifications

• H04L41/0896 Primary

  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

• H04L41/0895

  Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements · CPC title

• G06F2009/45579

  I/O management, e.g. providing access to device drivers or storage · CPC title

• G06F9/50

  Allocation of resources, e.g. of the central processing unit [CPU] · CPC title

• G06F9/45558

  Hypervisor-specific management and integration aspects · CPC title