Prescriptive cloud computing resource sizing based on multi-stream data sources

Patent metadata

Field	Value
Publication number	US-10459757-B1
Application number	US-201916411064-A
Country	US
Kind code	B1
Filing date	May 13, 2019
Priority date	May 13, 2019
Publication date	Oct 29, 2019
Grant date	Oct 29, 2019

Abstract

A multi-layer compute sizing correction stack may generate prescriptive compute sizing correction tokens for controlling sizing adjustments for computing resources. The input layer of the compute sizing correction stack may generate cleansed utilization data based on historical utilization data received via network connection. The input layer may receive one or more resource configurations that may be applied to implement the sizing correction. A prescriptive engine layer may generate a compute sizing correction trajectory indicative of a sizing adjustment to a computing resource. The compute sizing correction trajectory may account of historic processor, network, and memory utilization. Based on the compute sizing correction trajectory and a selected resource configuration, the prescriptive engine layer may generate the compute sizing correction tokens that that may be used to control compute sizing adjustments prescriptively.

First claim

What is claimed is:

1. A system including:
network interface circuitry configured to:
receive historical utilization data and tagging data for a selected virtual machine;
receive computer resource configurations;
send a compute sizing correction (CSC) token to a host interface, the host interface configured to control requisition for the selected virtual machine;

sizing circuitry in data communication with the network interface circuitry, the sizing circuitry configured to execute a CSC stack,
the CSC stack including:
a data staging layer;
an input layer;
a configuration layer; and
a prescriptive engine layer;

the CSC stack executable to:
obtain, via the input layer, the historical utilization data;
process, at the input layer, the historical utilization data to generate cleansed utilization data, the cleansed utilization data comprising a network utilization metric, a memory utilization metric, and a processor utilization metric;
store, at the data staging layer, the cleansed utilization data and the tagging data of the selected virtual machine;
determine, at the configuration layer, a compute utilization sizing criterion comprising logic to generate to a CSC trajectory based on the network utilization metric, the memory utilization metric, and the processor utilization metric;
determine, at the configuration layer, a resource transition criteria;
store, at the data staging layer, the compute utilization sizing criterion and the resource transition criteria;
access, at the prescriptive engine layer, the cleansed utilization data, and the tagging data of the selected virtual machine and the resource transition criteria via a memory resource provided by the data staging layer;
based on the network utilization metric, the memory utilization metric, the processor utilization metric and the compute utilization sizing criterion, generate, at the prescriptive engine layer, the CSC trajectory for the selected virtual machine in response to satisfaction of a first condition, satisfaction of a second condition, and satisfaction of a third condition, the first condition being based on a processor utilization measurement, the second condition being based on a memory utilization measurement, and the third condition being based on a network utilization measurement;
based on the tagging data of the selected virtual machine and the resource transition criteria, select, at the prescriptive engine layer, a resource configuration from the resource configurations;
based on the CSC trajectory and the selected resource configuration, determine a CSC adjustment for the selected virtual machine; and
based on the CSC adjustment, generate the CSC token.

2. The system of claim 1 , wherein the CSC stack is configured to:
determine, based on the processor utilization metric, the processor utilization measurement, the processor utilization measurement comprising a peak processor utilization measurement over a time period, a percentile-based processor utilization measurement over the time period, or a combination thereof;
determine, based on the memory utilization metric, memory utilization measurement, the memory utilization measurement comprising a peak memory utilization measurement over the time period, percentile based utilization measurement over the time period, or a combination thereof; and
determine, based on the network utilization metric, the network utilization measurement, the network utilization measurement comprising a peak network traffic measurement, a percent-based network traffic measurement, or a combination thereof.

3. The system of claim 2 ,
wherein satisfaction of the first condition comprises the peak processor utilization measurement being greater than a threshold processor utilization value and the percentile-based processor utilization measurement being less than a percentile based processor threshold value,
wherein satisfaction of the second condition comprises the peak memory utilization measurement being greater than a threshold memory utilization value and the percentile-based memory utilization measurement being less than a percentile-based memory threshold value, and
wherein satisfaction of the third condition comprises a network traffic ratio being greater than a network traffic utilization threshold, wherein the network traffic ratio comprises a ratio of the peak network traffic measurement and the percentile based traffic measurement.

4. The system of claim 1 , wherein the CSC stack is further configured to:
determine, based on the resource transition criteria, the tagging data of the selected virtual machine is compatible with the resource configuration, wherein the resource configuration is selected in response to determination that the tagging data of the selected virtual machine is compatible with the resource configuration.

5. The system of claim 4 , wherein to determine, based on the resource transition criteria, the tagging data of the selected virtual machine is compatible with the at least one of the resource configurations, the CSC stack is further configured to:
identify, from the resource configurations, the resource configuration in response to predetermined tagging data of the resource configuration matching the tagging data of the selected virtual machine.

6. The system of claim 1 , wherein the CSC token, when sent to the host interface, causes the host interface to implement the CSC adjustment by re-provisioning the selected virtual machine based on the selected resource configuration.

7. The system of claim 1 , wherein the CSC stack is further executable to determine the CSC token based on a feedback history generated using previous command inputs received from a CSC control interface generated at a presentation layer of the CSC stack.

8. The system of claim 1 , wherein the CSC stack is further executable to, at a presentation layer, generate CSC control interface including a CSC-window presentation, wherein the CSC-window presentation includes a selectable option to implement the CSC adjustment.

9. The system of claim 8 , wherein the CSC adjustment is grouped within the CSC-window presentation with other CSC adjustments corresponding to consumption savings within a pre-defined range.

10. The system of claim 8 , wherein the CSC-window presentation includes a summary table detailing CSC adjustments for multiple virtual machines.

11. The system of claim 1 , further the CSC stack executable to:
receive supplemental tagging data for at least one of the computer resource configurations;
enrich the computer resource configurations with the supplemental tagging data; and
rank the computer resource configurations based on the supplemental tagging data, wherein to select to resource configuration from the resource configurations, CSC stack is further executable to:
select the resource configuration in response to the resource configuration being highest ranked among the resource configurations.

12. The system of claim 1 , wherein the CSC token is configured to alter a compute capability for the selected virtual machine while preserving a selected operating system, region, networking throughput, or any combination thereof for the selected virtual machine.

13. A method including:
at network interface circuitry:
receiving historical utilization data for a selected virtual machine;
receiving tagging data for the selected virtual machine; and
receiving computer resource configuration data;

at sizing circuitry in data communication with the network interface circuitry, the sizing circuitry executing a compute sizing correction (CSC) stack:
obtaining, via an input layer of the CSC stack, the h

Assignees

• Accenture Global Solutions Ltd

Inventors

• Srinivasan Madhan Kumar
• Pv Guruprasad
• Purushothaman Arun

Classifications

• G06F2009/45583

  Memory management, e.g. access or allocation · CPC title

• G06F2009/45595

  Network integration; Enabling network access in virtual machine instances · CPC title

• G06F2009/45591

  Monitoring or debugging support · CPC title

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title

• G06F9/5072 Primary

  Grid computing · CPC title