Method of controlling access to computing resource within shared computing environment

• US 7,827,361 B1
• Filed: 10/21/2004
• Issued: 11/02/2010
• Est. Priority Date: 10/21/2004
• Status: Expired due to Fees

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Karlsson, Magnus, Karamanolis, Christos
• Primary Examiner(s): Bragdon; Reginald G
• Assistant Examiner(s): Ruiz; Aracelis
• Application Number:
  US10/970,888
• Time in Patent Office: 2,203 Days
• Field of Search: 711/147, 709/212, 709/221, 709/226, 709/229, 709/250, 718/105
• US Class Current: 711/147
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system

Abstract

An embodiment of a method of controlling access to a computing resource within a shared computing environment begins with a first step of determining a plurality of controller functions for a plurality of operating ranges for workloads accessing the computing resource. Each of the controller functions comprises a mathematical operator which takes an input and provides an output. The method continues by iteratively performing second through fifth steps. In the second step, the method measures performance parameters for the workloads to determine a performance parameter vector for the workloads. In the third step, the method compares the performance parameter vector to a reference performance parameter vector to determine an error parameter. In the fourth step, the method applies a particular controller function selected from the plurality of controller functions to the error parameter to determine a target throughput for each of the workloads. In the fourth step, the method adjusts access to the computing resource for each of the workloads having a throughput limit different from about the target throughput for the workload by reducing or increasing the throughput limit for the workload to about the target throughput.

Claims

1. A method of controlling access to a computing resource within a shared computing environment comprising:

determining, by a computer, a plurality of controller functions for a plurality of operating ranges for workloads accessing the computing resource, each of the controller functions comprising a mathematical operator which takes an input and provides an output;

iteratively performing by the computer;

measuring performance parameters for the workloads to determine a performance parameter vector for the workloads;

comparing the performance parameter vector to a reference performance parameter vector to determine an error parameter;

applying a particular controller function selected from the plurality of controller functions to the error parameter to determine a target throughput for each of the workloads; and

for each workload having a throughput limit different from the target throughput for the workload, adjusting access to the computing resource by the workload by reducing or increasing the throughput limit for the workload based on the target throughput.

2. The method of claim 1 wherein determining the plurality of controller functions comprises performing a regression analysis on white noise input and resulting latencies.

3. The method of claim 1 wherein the workloads comprise desired input/output operations for storage.

4. The method of claim 1 wherein the controller functions comprise a plurality of linear controller functions.

5. The method of claim 1 wherein the performance parameters comprise latencies.

6. The method of claim 1 wherein applying the particular controller function to the error parameter comprises applying the particular controller function that also takes into account a prior target throughput.

7. The method of claim 1 wherein adjusting the access to the computing resource comprises adjusting a throttling module to reduce or increase the throughput.

8. The method of claim 1 wherein the error parameter is determined by subtracting the performance parameter vector from the reference performance parameter vector which forms an error vector and selecting a minimum term from the error vector as the error parameter.

9. The method of claim 1 wherein applying the particular controller function to the error parameter further comprises applying the controller function to a current aggregate target throughput.

10. The method of claim 1 wherein adjusting the access to the computing resource for the workload employs a throughput allocation table which specifies shares for the workloads according to throughput bands.

11. The method of claim 1 wherein the computing resource is selected from a group consisting of storage, processor cycles, processing units, network bandwidth, memory, servers, another computing resource, and a combination thereof.

12. The method of claim 1, wherein reducing or increasing the throughput limit based on the target throughput comprises reducing or increasing the throughput limit to about the target throughput.

14. The method of claim 1 wherein the error parameter is determined by subtracting the latency vector from the reference latency vector which forms an error vector and selecting a minimum term from the error vector as the error parameter.

15. The method of claim 1 wherein applying the controller function to the error parameter further comprises applying the controller function to a current aggregate target throughput.

16. The method of claim 1 wherein adjusting the access to the storage system for the workload employs a throughput allocation table which specifies shares for the workloads according to throughput bands.

13. A method of controlling access to a storage system comprising the steps of:

determining, by a computer, a plurality of controller functions for a plurality of operating ranges for workloads accessing the storage system, each of the controller functions comprising a mathematical operator which takes an input and provides an output;

iteratively performing by the computer;

measuring latencies for the workloads to determine a latency vector for the clients;

comparing the latency vector to a reference latency vector to determine an error parameter

applying a particular controller function selected from the plurality of controller functions to the error parameter to determine a target throughput for each of the workloads; and

for each workload having a throughput limit different from the target throughput for the workload, adjusting access to the storage system by the workload by reducing or increasing the target throughput for the workload based on the target throughput.

17. The method of claim 13, wherein reducing or increasing the throughput limit based on the target throughput comprises reducing or increasing the throughput limit to about the target throughput.

18. A computer readable media comprising computer code for implementing a method of controlling access to a computing resource within a shared computing environment, the method of controlling access to the computing resource comprising the steps of:

determining a plurality of controller functions for a plurality of operating ranges for workloads accessing the computing resource, each of the controller functions comprising a mathematical operator which takes an input and provides an output;

iteratively performing the steps of;

measuring performance parameters for the workloads to determine a performance parameter vector for the workloads;

comparing the performance parameter vector to a reference performance parameter vector to determine an error parameter;

applying a particular controller function selected from the plurality of controller functions to the error parameter to determine a target throughput for each of the workloads; and

for each workload having a throughput limit different from about the target throughput for the workload, adjusting access to the computing resource by the workload by reducing or increasing the throughput limit for the workload to about the target throughput.

19. A computer readable media comprising computer code for implementing a method of controlling access to a computing resource within a shared computing environment, the method of controlling access to the computing resource comprising the steps of:

determining a plurality of controller functions for a plurality of operating ranges for workloads accessing the storage system, each of the controller functions comprising a mathematical operator which takes an input and provides an output;

iteratively performing the steps of;

measuring latencies for the workloads to determine a latency vector for the clients;

comparing the latency vector to a reference latency vector to determine an error parameter

applying a particular controller function selected from the plurality of controller functions to the error parameter to determine a target throughput for each of the workloads; and

for each workload having a throughput limit different from the target throughput for the workload, adjusting access to the storage system by the workload by reducing or increasing the target throughput for the workload based on the target throughput.

20. The computer readable media of claim 19, wherein reducing or increasing the throughput limit based on the target throughput comprises reducing or increasing the throughput limit to about the target throughput.