Adaptive idle detection in a software-defined data center in a hyper-converged infrastructure

Patent metadata

Field	Value
Publication number	US-12067430-B2
Application number	US-202117468701-A
Country	US
Kind code	B2
Filing date	Sep 8, 2021
Priority date	Jul 22, 2021
Publication date	Aug 20, 2024
Grant date	Aug 20, 2024

Abstract

An adaptive idle detection method determines whether software defined data centers (SDDCs) in a hyperconverged infrastructure (HCI) environment are idle. Idleness may be quantified via a coefficient of variation (CV) against resource usage, so as to adapt the idle detection method to SDDCs with different hardware specifications and workloads. Management overhead may also be filtered out by the idle detection method, and the idle detection method may use idleness scores to further reduce overhead.

First claim

We claim:

1. A method to detect idleness of a software-defined data center (SDDC), the method comprising:
for each host of a plurality of hosts in the SDDC:
obtaining resource usage samples provided by the host;
determining a coefficient of variation (CV) associated with fluctuations of the resource usage samples;
determining an idleness score of the host based at least in part on the usage samples and the CV;
wherein determining the idleness score of the host comprises:
determining a first idleness score associated with processor usage by the host;
determining a second idleness score associated with network usage by the host; and
determining the idleness score of the host from a smallest of either the first idleness score or the second idleness score; and

based at least in part on the idleness score, determining that the host is idle;

determining that the SDDC is idle, in response to determination that all of the hosts of the plurality of hosts are idle; and
triggering an event to perform a remediation action for the idle SDDC.

2. The method of claim 1 , further comprising filtering outlier samples associated with a management overhead, prior to determining the idleness score of the hosts.

3. The method of claim 1 , wherein determining that the host is idle includes:
determining that the idleness score exceeds a candidate score, so as to qualify the host for further determination of idleness; and
in response to determination that the idleness score exceeds the candidate score, determining that the idleness score exceeds a graduate score, wherein the host is determined to be idle if the idleness score exceeds the graduate score.

4. The method of claim 3 , wherein determining that the idleness score exceeds a graduate score comprises:
fetching resource usage samples from a database;
and calculating the graduated score using the fetched resource usage samples.

5. The method of claim 1 , further comprising determining that the SDDC is in use, in response to determination that at least one host of the plurality of hosts is in use.

6. The method of claim 1 , wherein determining the idleness score reduces overhead caused by retrieval of resource usage samples from a usage statistics database.

7. The method of claim 1 , further comprising:
determining that the host is in use, in response to the idleness score of the host being less than a candidate score, wherein the candidate score is calculated based on an interval at which a resource usage sample is collected, a time span during which the host is detected, and an interval at which a detection cycle occurs; and
determining that the host is in use, in response to the idleness score of the host being less than a graduate score, wherein the graduate score is calculated based on the time span during which the host is detected and the interval at which the resource usage sample is collected.

8. A non-transitory computer-readable medium having instructions stored thereon, which in response to execution by one or more processors, cause the one or more processors of a host to perform or control performance of a method to detect idleness of a software-defined data center (SDDC), wherein the method comprises:
for each host of a plurality of hosts in the SDDC:
obtaining resource usage samples provided by the host;
determining a coefficient of variation (CV) associated with fluctuations of the resource usage samples;
determining an idleness score of the host based at least in part on the usage samples and the CV;
wherein determining the idleness score of the host comprises:
determining a first idleness score associated with processor usage by the host;
determining a second idleness score associated with network usage by the host; and
determining the idleness score of the host from a smallest of either the first idleness score or the second idleness score; and

based at least in part on the idleness score, determining that the host is idle;

determining that the SDDC is idle, in response to determination that all of the hosts of the plurality of hosts are idle; and
triggering an event to perform a remediation action for the idle SDDC.

9. The non-transitory computer-readable medium of claim 8 , wherein the method further comprises:
filtering outlier samples associated with a management overhead, prior to determining the idleness score of the hosts.

10. The non-transitory computer-readable medium of claim 8 , wherein determining that the host is idle includes:
determining that the idleness score exceeds a candidate score, so as to qualify the host for further determination of idleness; and
in response to determination that the idleness score exceeds the candidate score, determining that the idleness score exceeds a graduate score, wherein the host is determined to be idle if the idleness score exceeds the graduate score.

11. The non-transitory computer-readable medium of claim 10 , wherein determining that the idleness score exceeds a graduate score comprises:
fetching resource usage samples from a database;
and calculating the graduated score using the fetched resource usage samples.

12. The non-transitory computer-readable medium of claim 8 , wherein the method further comprises:
determining that the SDDC is in use, in response to determination that at least one host of the plurality of hosts is in use.

13. The non-transitory computer-readable medium of claim 8 , wherein determining the idleness score reduces overhead caused by retrieval of resource usage samples from a usage statistics database.

14. The non-transitory computer-readable medium of claim 8 , wherein the method further comprises:
determining that the host is in use, in response to the idleness score of the host being less than a candidate score, wherein the candidate score is calculated based on an interval at which a resource usage sample is collected, a time span during which the host is detected, and an interval at which a detection cycle occurs; and
determining that the host is in use, in response to the idleness score of the host being less than a graduate score, wherein the graduate score is calculated based on the time span during which the host is detected and the interval at which the resource usage sample is collected.

15. A network device to detect idleness of a software-defined data center (SDDC), the network device comprising:
one or more processors; and
a non-transitory computer-readable medium coupled to the one or more processors, and having instructions stored thereon, which in response to execution by the one or more processors, cause the one or more processors to perform or control performance of operations that include:
for each host of a plurality of hosts in the SDDC:
obtain resource usage samples provided by the host;
determine a coefficient of variation (CV) associated with fluctuations of the resource usage samples;
determine an idleness score of the host based at least in part on the usage samples and the CV;
wherein determining the idleness score of the host comprises:
determining a first idleness score associated with processor usage by the host;
determining a second idleness score associated with network usage by the host; and
determining the idleness score of the host from a smallest of either the first idleness score or the second idleness score; and

based at least in part on the idleness score, determine that the host is idle;

determine that the SDDC is idle, in response to determination that all of the hosts of the plurality of hosts are idle; and
trigger an event to perform a remediation action for the idle SDD

Assignees

• Vmware Inc
• VMware LLC

Inventors

• Wan Jiushi
• Feng Jin
• Huang Zhou
• Zhao Jian
• Yang Yang

Classifications

• G06F9/5005

  to service a request · CPC title

• G06F2209/508

  Monitor · CPC title

• G06F9/542 Primary

  Event management; Broadcasting; Multicasting; Notifications · CPC title

• G06F9/5077 Primary

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title