Managing client computing systems using distilled data streams

Patent metadata

Field	Value
Publication number	US-11265398-B2
Application number	US-202017135390-A
Country	US
Kind code	B2
Filing date	Dec 28, 2020
Priority date	Jan 24, 2019
Publication date	Mar 1, 2022
Grant date	Mar 1, 2022

Abstract

The present disclosure relates to techniques for managing client computing systems, such as a client distributed-computing system. In one embodiment, a desired state of the client distributed-computing system and a current state of the client distributed-computing system are received. Measurement data generated by a plurality of components of the client distributed-computing system is received. The measurement data is processed using one or more analytical or statistical techniques to generate distilled data. Based on one or more policies and the desired state, the distilled data and the current state are analyzed to determine one or more actions predicted to converge the current state of the client distributed-computing system towards the desired state. One or more control commands corresponding to the one or more actions are provided to the client distributed-computing system. The one or more control commands cause the client distributed-computing system to perform the one or more actions.

First claim

What is claimed is:

1. A method for managing, by a cloud-services distributed-computing system, a client distributed-computing system, the method comprising:
at the cloud-services distributed-computing system:
receiving a desired state of the client distributed-computing system, the desired state defined according to user input;
receiving a current state of the client distributed-computing system;
receiving measurement data generated by a plurality of components of the client distributed-computing system;
generating distilled data by processing the measurement data using one or more analytical or statistical techniques based on previous state changes of the client distributed-computing system resulting from control commands issued prior to receiving the current state of the client distributed-computing system;
based on one or more policies and the desired state, analyzing the distilled data and the current state to determine one or more actions predicted to converge the current state of the client distributed-computing system towards the desired state; and
providing one or more control commands to the client distributed-computing system, wherein the one or more control commands cause the client distributed-computing system to perform the one or more actions.

2. The method of claim 1 , wherein processing the measurement data using the one or more analytical or statistical techniques comprises filtering the measurement data to identify a subset of the measurement data, wherein the distilled data includes the identified subset of the measurement data.

3. The method of claim 1 , wherein processing the measurement data using the one or more analytical or statistical techniques comprises applying one or more functions to at least a portion of the measurement data to generate at least a portion of the distilled data.

4. The method of claim 3 , wherein the one or more functions include a derivative function.

5. The method of claim 1 , wherein the one or more analytical or statistical techniques include machine learning techniques.

6. The method of claim 1 , wherein processing the measurement data using the one or more analytical or statistical techniques comprises generating statistical data from the measurement data, wherein the distilled data includes the generated statistical data.

7. The method of claim 1 , wherein determining the one or more actions further comprises:
determining, based on the current state and the one or more policies, a plurality of scores for a plurality of predefined actions, wherein the one or more actions are selected from the plurality of predefined actions based on the plurality of scores.

8. The method of claim 1 , further comprising:
receiving a subsequent state of the client distributed-computing system, the subsequent state generated by the client distributed-computing system after performing the one or more actions; and
updating the one or more policies based on a difference between the subsequent state and the current state relative to the desired state.

9. The method of claim 1 , wherein the one or more actions include increasing or decreasing a resolution of monitoring for one or more parameters of the client distributed-computing system.

10. The method of claim 1 , wherein the measurement data includes performance metrics for the plurality of components of the client distributed-computing system.

11. The method of claim 1 , wherein the measurement data includes sensor data.

12. The method of claim 1 , wherein the measurement data is received from the client distributed-computing system via a plurality of data streams, and wherein each data stream of the plurality of data streams is assigned to a respective component of the plurality of components.

13. The method of claim 1 , further comprising:
prior to receiving the current state and the measurement data, providing one or more previous control commands to the client distributed-computing system, wherein the one or more previous control commands cause the client distributed-computing system to perform the one or more previous actions and to change from a previous state to the current state, and wherein the measurement data is received within a predefined time period after providing the one or more previous control commands.

14. The method of claim 13 , wherein the one or more actions are determined further based on a difference between the current state and the previous state.

15. A non-transitory computer-readable storage medium storing one or more programs configured to be executed by one or more processors of a cloud-services distributed-computing system, wherein the one or more programs include instructions for:
processing a received desired state of a client distributed-computing system, the desired state defined according to user input;
processing a received current state of the client distributed-computing system, the current state generated by the client distributed-computing system;
processing received measurement data generated by a plurality of components of the client distributed-computing system;
generating distilled data by processing the measurement data using one or more analytical or statistical techniques based on previous state changes of the client distributed-computing system resulting from control commands issued prior to receiving the current state of the client distributed-computing system;
based on one or more predefined policies and the desired state, analyzing the distilled data and the current state to determine one or more actions predicted to converge the current state of the client distributed-computing system towards the desired state; and
providing one or more control commands to the client distributed-computing system, wherein the one or more control commands cause the client distributed-computing system to perform the one or more actions.

16. A cloud-services distributed-computing system, comprising:
one or more processors; and
memory storing one or more programs configured to be executed by the one or more processors, the one or more programs including instructions for:
processing a received desired state of a client distributed-computing system, the desired state defined according to user input;
processing a received current state of the client distributed-computing system, the current state generated by the client distributed-computing system;
processing received measurement data generated by a plurality of components of the client distributed-computing system;
generating distilled data by processing the measurement data using one or more analytical or statistical techniques based on previous state changes of the client distributed-computing system resulting from control commands issued prior to receiving the current state of the client distributed-computing system;
based on one or more predefined policies and the desired state, analyzing the distilled data and the current state to determine one or more actions predicted to converge the current state of the client distributed-computing system towards the desired state; and
providing one or more control commands to the client distributed-computing system, wherein the one or more control commands cause the client distributed-computing system to perform the one or more actions.

17. The cloud-services distributed-computing system of claim 16 , wherein the past observations correspond to respective past actions performed by the client computing system in response to previous commands performed at the cloud-services distributed-computing system.

Assignees

• Vmware Inc

Inventors

• Cooley James W
• Dar Sufian A

Classifications

• H04L67/1097

  for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS] · CPC title

• H04L41/14

  Network analysis or design · CPC title

• H04L41/0895

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

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title

• H04L67/34 Primary

  involving the movement of software or configuration parameters  (network booting or remote initial program loading [RIPL] G06F9/4416) · CPC title