System and method for sharing central processing unit (CPU) resources with unbalanced applications

Patent metadata

Field	Value
Publication number	US-11494236-B2
Application number	US-202016867613-A
Country	US
Kind code	B2
Filing date	May 6, 2020
Priority date	May 6, 2020
Publication date	Nov 8, 2022
Grant date	Nov 8, 2022

Abstract

A method, computer program product, and computing system for monitoring utilization of each central processing unit (CPU) core of a plurality of CPU cores. An average input/output (IO) latency for an operating system thread executing on the CPU core of the plurality of CPU cores may be determined. The operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core may be adjusted based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores.

First claim

What is claimed is:

1. A computer-implemented method, executed on a computing device, comprising:
monitoring, via the computing device, utilization of each central processing unit (CPU) core of a plurality of CPU cores, wherein monitoring utilization of each CPU core of a plurality of CPU cores includes: monitoring a block application utilization and monitoring a file application utilization;
determining an average input/output (IO) latency for an operating system thread executing on each CPU core of the plurality of CPU cores;
adjusting an operating system thread IO polling cadence for at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores, wherein adjusting the operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores includes:
determining whether the average IO latency for the operating system thread executing on the at least one CPU core is greater than average IO latencies of other CPU cores of the plurality of CPU cores by more than a predefined IO latency threshold;
determining whether the file application utilization of the at least one CPU core is greater than a predefined file application utilization threshold; and
preventing at least one new block application IO request from being fetched via the operating system thread executing on the at least one CPU core of the plurality of CPU cores in response to determining that the file application utilization of the CPU core is greater than the predefined file application utilization threshold.

2. The computer-implemented method of claim 1 , adjusting the operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores includes:
determining whether an average block application utilization of the plurality of CPU cores is less than a first predefined block application utilization threshold.

3. The computer-implemented method of claim 2 , wherein adjusting the operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores includes:
decreasing the operating system thread IO polling cadence for the at least one operating system thread executing on the at least one CPU core in response to determining that the average IO latency for the operating system thread executing on the CPU core is greater than the average IO latencies of the other CPU cores of the plurality of CPU cores by more than a predefined IO latency threshold and that the average block application utilization of the plurality of CPU cores is less than the predefined block application utilization threshold.

4. The computer-implemented method of claim 3 , wherein adjusting the operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores includes:
determining whether the block application utilization of one or more CPU cores is greater than a second predefined block application utilization threshold; and
increasing the operating system thread IO polling cadence for each operating system thread executing on one or more other CPU cores in response to determining that the block application utilization of the one or more CPU cores is greater than the second predefined block application utilization threshold.

5. A computer program product residing on a non-transitory computer readable medium having a plurality of instructions stored thereon which, when executed by a processor, cause the processor to perform operations comprising:
monitoring utilization of each central processing unit (CPU) core of a plurality of CPU cores, wherein monitoring utilization of each CPU core of a plurality of CPU cores includes: monitoring a block application utilization and monitoring a file application utilization;
determining an average input/output (IO) latency for an operating system thread executing on the CPU core of the plurality of CPU cores;
adjusting an operating system thread IO polling cadence for at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores, wherein adjusting the operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores includes:
determining whether the average IO latency for the operating system thread executing on the at least one CPU core is greater than average IO latencies of other CPU cores of the plurality of CPU cores by more than a predefined IO latency threshold;
determining whether the file application utilization of the at least one CPU core is greater than a predefined file application utilization threshold; and
preventing at least one new block application IO request from being fetched via the operating system thread executing on the at least one CPU core of the plurality of CPU cores in response to determining that the file application utilization of the CPU core is greater than the predefined file application utilization threshold.

6. The computer program product of claim 5 , wherein adjusting the operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores includes:
determining whether an average block application utilization of the plurality of CPU cores is less than a first predefined block application utilization threshold.

7. The computer program product of claim 6 , wherein adjusting the operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core based upon, at least in part, the utilization of each CPU core of the plurality of CPU cores and the average IO latency for the operating system thread executing on each CPU core of the plurality of CPU cores includes:
decreasing the operating system thread IO polling cadence for the at least one operating system thread executing on at least one CPU core in response to determining that the average IO latency for the operating system thread executing on the CPU core is greater than the average IO latencies of the other CPU cores of the plurality of CPU cores by more than a predefined IO latency threshold

Assignees

• Emc Ip Holding Co Llc
• EMP IP Holding Company LLC

Inventors

• Kamran Lior
• Alkalay Amitai

Classifications

• G06F9/5027 Primary

  the resource being a machine, e.g. CPUs, Servers, Terminals · CPC title

• G06F11/3024

  where the computing system component is a central processing unit [CPU] · CPC title

• G06F2209/5018

  Thread allocation · CPC title

• G06F11/3006

  where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems (multiprogramming arrangements G06F9/46; allocation of resources G06F9/50) · CPC title

• G06F9/4881

  Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues · CPC title