ADAPTIVE BANDWIDTH DISTRIBUTION SYSTEM FOR HIGH-PERFORMANCE INPUT/OUTPUT DEVICES WITH VARIABLE THROUGHPUT

• US 20090271531A1
• Filed: 07/24/2008
• Published: 10/29/2009
• Est. Priority Date: 04/23/2008
• Status: Active Grant

• Current Assignee: Autodesk, Inc.
• Original Assignee: Autodesk, Inc.
• Inventors: Vinet, Eric, Labute, Daniel
• Granted Patent: US 8,060,660 B2
• Time in Patent Office: Days
• US Class Current: 710/6
• CPC Class Codes: G06F 13/22 using successive scanning, e.g. polling G06F13/24 takes precedence
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/0671 In-line storage system

Abstract

A method for issuing shadow requests to manage bandwidth allocation between an application that issues input/output (I/O) operation requests and an I/O device. A bandwidth manager detects the completion of an I/O operation, which includes either a read operation or a write operation. The bandwidth manager calculates a statistical duration for future I/O operations between the application and the I/O device based on throughput statistics related to past I/O operations. The bandwidth manager generates a shadow request for reserving a position in a queue that stores pending I/O requests for the I/O device for a first future I/O operation request from the application and having a duration related to the statistical duration, and inserts the shadow request into the queue. Advantageously, applications that do not make frequent I/O operation requests in advance may still execute I/O operations because bandwidth is reserved for future I/O operation requests via the shadow requests.

Claims

1. A method for processing input/output (I/O) requests located within a queue for managing bandwidth allocation between an application that issues I/O operation requests and an I/O device, the method comprising:

detecting that an I/O request is at a front of a queue of pending I/O requests associated with the I/O device;

determining that the I/O request is a shadow request, wherein the shadow request reserves a position in the queue for a future I/O operation request from the application, and wherein the shadow request is associated with a duration of time;

polling the application for the duration of time to determine whether the application issues an I/O operation request that includes a request for either of a read operation from the I/O device to the application or a write operation from the application to the I/O device;

receiving a first I/O operation request from the application during the duration of time associated with the shadow request; and

unblocking the application to allow a first I/O operation associated with the first I/O operation request to be executed.

2. The method of claim 1, further comprising:

executing the I/O operation between the application and the I/O device; and

updating throughput statistics related to past I/O operations between the application and the I/O device.

3. The method of claim 2, wherein the throughput statistics include one or more values corresponding to a statistical duration associated with previously executed I/O operations between the application and the I/O device.

4. The method of claim 3, wherein the throughput statistics include one or more values corresponding to a data size associated with previously executed I/O operations executed between the application and the I/O device.

5. The method of claim 4, further comprising the step of calculating an I/O operation request frequency reflecting the rate at which the application issues I/O operation requests to the I/O device based on an average interval between timestamps recorded when the application issues I/O operation requests.

6. The method of claim 1, further comprising:

determining that the I/O request at the front of the queue is an actual I/O request;

executing a second I/O operation associated with the actual I/O request between the application and the I/O device; and

updating throughput statistics related to past I/O operations between the application and the I/O device.

7. The method of claim 6, wherein the throughput statistics include one or more values corresponding to a statistical duration associated with previously executed I/O operations between the application and the I/O device.

8. The method of claim 7, wherein the throughput statistics include one or more values corresponding to a data size associated with previously executed I/O operations executed between the application and the I/O device.

9. The method of claim 8, further comprising the step of calculating an I/O operation request frequency reflecting the rate at which the application issues I/O operation requests to the I/O device based on an average interval between timestamps recorded when the application issues I/O operation requests.

10. The method of claim 1, wherein the I/O device comprises a hard disk, a memory, a computer monitor, or a disc drive.

11. A computer-readable medium storing instructions that, when executed by a processor, cause a bandwidth manager to process input/output (I/O) requests located within a queue for managing bandwidth allocation between an application that issues I/O operation requests and an I/O device, by performing the steps of:

detecting that an I/O request is at a front of a queue of pending I/O requests associated with the I/O device;

determining that the I/O request is a shadow request, wherein the shadow request reserves a position in the queue for a future I/O operation request from the application, and wherein the shadow request is associated with a duration of time;

polling the application for the duration of time to determine whether the application issues an I/O operation request that includes a request for either of a read operation from the I/O device to the application or a write operation from the application to the I/O device;

receiving a first I/O operation request from the application during the duration of time associated with the shadow request; and

unblocking the application to allow a first I/O operation associated with the first I/O operation request to be executed.

12. The computer-readable medium of claim 11, further comprising:

executing the I/O operation between the application and the I/O device; and

updating throughput statistics related to past I/O operations between the application and the I/O device.

13. The computer-readable medium of claim 12, wherein the throughput statistics include one or more values corresponding to a statistical duration associated with previously executed I/O operations between the application and the I/O device.

14. The computer-readable medium of claim 13, wherein the throughput statistics include one or more values corresponding to a data size associated with previously executed I/O operations executed between the application and the I/O device.

15. The computer-readable medium of claim 14, further comprising the step of calculating an I/O operation request frequency reflecting the rate at which the application issues I/O operation requests to the I/O device based on an average interval between timestamps recorded when the application issues I/O operation requests.

16. The computer-readable medium of claim 11, further comprising:

determining that the I/O request at the front of the queue is an actual I/O request;

executing a second I/O operation associated with the actual I/O request between the application and the I/O device; and

updating throughput statistics related to past I/O operations between the application and the I/O device.

17. The computer-readable medium of claim 16, wherein the throughput statistics include one or more values corresponding to a statistical duration associated with previously executed I/O operations between the application and the I/O device.

18. The computer-readable medium of claim 17, wherein the throughput statistics include one or more values corresponding to a data size associated with previously executed I/O operations executed between the application and the I/O device.

19. The computer-readable medium of claim 18, further comprising the step of calculating an I/O operation request frequency reflecting the rate at which the application issues I/O operation requests to the I/O device based on an average interval between timestamps recorded when the application issues I/O operation requests.

20. The computer-readable medium of claim 11, wherein the I/O device comprises a hard disk, a memory, a computer monitor, or a disc drive.