Approach for managing interrupt load distribution

• US 20070043970A1
• Filed: 08/03/2006
• Published: 02/22/2007
• Est. Priority Date: 08/22/2005
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Oracle America, Inc. (Oracle Corporation)
• Inventors: Tripathi, Sunay, Solomita, Ethan
• Granted Patent: US 7,610,425 B2
• Time in Patent Office: Days
• US Class Current: 714/10
• CPC Class Codes: G06F 11/00 Error detection; Error correction; Monitoring error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00
• CPC Class Codes: G06F 11/3024 where the computing system component is a central processing unit [CPU]
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 13/24 using interrupt G06F13/32 takes precedence
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/4812 by interrupt, e.g. masked
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system

Abstract

A method and apparatus for distributing multiple interrupts among multiple processors is disclosed. According to one embodiment, an interrupt daemon monitors the interrupt load among the processors that results from an initial mapping of the interrupts to the processors. The interrupt daemon determines whether there is a sufficient imbalance of the interrupts among the processors. If so, the interrupt daemon triggers a reassignment routine that generates a new mapping of the interrupts among the processors, and if not, the interrupt daemon goes to sleep for a specified time period. If the new mapping produces a sufficient improvement in the distribution of interrupts among the processors, based on the same criteria used to detect the imbalance, the new mapping is used by the central hub for subsequent distribution of interrupts to the processors. However, if the new mapping does not provide a sufficient improvement, the original mapping continues to be used.

Claims

1. In a system comprising a plurality of processors, a machine-implemented method comprising:

receiving a first mapping of one or more interrupt channels to one or more processors, wherein the first mapping includes a mapping of a particular device'"'"'s interrupts to a plurality of processors;

determining that a particular device does not use enough interrupt channels to comply with the first mapping;

designating a particular processor of the one or more processors as a managing processor;

designating a particular interrupt channel of the particular device'"'"'s interrupt channels as a managed interrupt channel;

distributing interrupts sent over the managed interrupt channel to the managing processor;

determining one or more tasks to be performed for each interrupt sent using the managed interrupt channel; and

causing the managing processor to assign the one or more tasks to a set of one or more additional processors from the plurality of processors, wherein the one or more additional processors perform at least one or more tasks for each interrupt sent using the managed interrupt channel.

2. The method of claim 1, wherein causing the managing processor to assign the one or more tasks to a set of one or more additional processors further comprises:

determining that a particular task is associated with a particular task category, wherein a task category identifies one or more tasks having common characteristics; and

depending on the category for the particular task, assigning the particular task to a particular processor of the one or more additional processors.

3. The method of claim 1, wherein causing the managing processor to assign the one or more tasks to a set of one or more additional processors includes causing the managing processor to successively assign each task to the next available processor

4. The method of claim 1, wherein causing the managing processor to assign the one or more tasks to a set of one or more additional processors further comprises:

causing the managing processor to perform one or more operations on data representing a particular task, wherein performing the one or more operations causes an output to be generated; and

based upon the output of the one or more operations, assigning the particular task to particular processor of the one or more additional processors.

5. The method of claim 1, wherein the particular device sends interrupts using a single interrupt channel, and wherein designating a particular interrupt channel as the managed interrupt channel includes designating the single interrupt channel as the managed interrupt channel.

6. The method of claim 1, wherein determining that the particular device does not use enough interrupt channels to comply with the first mapping includes:

determining that the particular device uses a particular amount of interrupt channels to send one or more interrupts;

determining that the first mapping requires the particular device'"'"'s interrupts to be distributed among a particular amount of processors; and

determining that the particular amount of processors is greater than the particular amount of interrupt channels.

7. The method of claim 1, wherein the managing processor and the one or more additional processors are in a set of one or more processors for the particular device, and wherein the first mapping requires the particular device'"'"'s interrupts to be distributed among processors identified in the set of one or more processors.

8. The method of claim 7, wherein at least one processor in the set of processors handles interrupts from only the particular device.

9. The method of claim 7, wherein the user designates at least one processor in the set of processors as reserved to handle interrupts from only the particular device.

10. The method of claim 1, wherein a user specifies a set of one or more processors for the particular device, and wherein the first mapping includes a mapping of the one or more interrupt channels to at least one processor in the set of processors specified by the user.

11. The method of claim 1, further comprising

receiving a second mapping of the one or more interrupt channels to the one or more processors;

and

distributing the one or more interrupts from the particular device among the one or more processors based on the second mapping.

12. The method of claim 11, wherein the second mapping is received in response to determining that the load from the particular device has exceed a threshold load for the one or more processors, and wherein the second mapping causes the processor load caused by the particular device to fall below the threshold.

13. The method of claim 12, further comprising:

determining that a particular device does not use enough interrupt channels to comply with the second mapping;

designating a particular interrupt channel of the second device as a managed interrupt channel;

selecting a new processor of the plurality of processors as the managing processor; and

distributing interrupts sent over the managed interrupt channel to the newly selected managing processor, wherein the newly selected managing processor assigns one or more tasks associated with interrupts from the particular device to a set of one or more additional processors from the plurality of processors.

14. A machine-readable medium comprising a set of instructions which, when executed by one or more processors, causes the one or more processors to perform the following operations:

receiving a first mapping of one or more interrupt channels to one or more processors, wherein the first mapping includes a mapping of a particular device'"'"'s interrupts to a plurality of processors;

determining that a particular device does not use enough interrupt channels to comply with the first mapping;

designating a particular processor of the one or more processors as a managing processor;

designating a particular interrupt channel of the particular device'"'"'s interrupt channels as a managed interrupt channel;

distributing interrupts sent over the managed interrupt channel to the managing processor;

determining one or more tasks to be performed for each interrupt sent using the managed interrupt channel; and

causing the managing processor to assign the one or more tasks to a set of one or more additional processors from the plurality of processors, wherein the one or more additional processors perform at least one or more tasks for each interrupt sent using the managed interrupt channel.

15. The machine-readable medium of claim 14, wherein causing the managing processor to assign the one or more tasks to a set of one or more additional processors includes:

determining that a particular task is associated with a particular task category, wherein a task category identifies one or more tasks having common characteristics; and

depending on the category for the particular task, assigning the particular task to a particular processor of the one or more additional processors.

16. The machine-readable medium of claim 14, wherein causing the managing processor to assign the one or more tasks to a set of one or more additional processors includes causing the managing processor to successively assign each task to the next available processor

17. The machine-readable medium of claim 14, wherein causing the managing processor to assign the one or more tasks to a set of one or more additional processors includes:

causing the managing processor to perform one or more operations on data representing a particular task, wherein performing the one or more operations causes an output to be generated; and

based upon the output of the one or more operations, assigning the particular task to particular processor of the one or more additional processors.

18. The machine-readable medium of claim 14, wherein the particular device sends interrupts using a single interrupt channel, and wherein designating a particular interrupt channel as the managed interrupt channel includes designating the single interrupt channel as the managed interrupt channel.

19. The machine-readable medium of claim 14, wherein determining that the particular device does not use enough interrupt channels to comply with the first mapping includes:

determining that the particular device uses a particular amount of interrupt channels to send one or more interrupts;

determining that the first mapping requires the particular device'"'"'s interrupts to be distributed among a particular amount of processors; and

determining that the particular amount of processors is greater than the particular amount of interrupt channels.

20. The machine-readable medium of claim 14, wherein the managing processor and the one or more additional processors are in a set of one or more processors for the particular device, and wherein the first mapping requires the particular device'"'"'s interrupts to be distributed among processors identified in the set of one or more processors.

21. The machine-readable medium of claim 20, wherein at least one processor in the set of processors handles interrupts from only the particular device.

22. The machine-readable medium of claim 20, wherein the user designates at least one processor in the set of processors as reserved to handle interrupts from only the particular device.

23. The machine-readable medium of claim 14, wherein a user specifies a set of one or more processors for the particular device, and wherein the first mapping includes a mapping of the one or more interrupt channels to at least one processor in the set of processors specified by the user.

24. The machine-readable medium of claim 14, wherein the set of instructions causes the one or more processors to further perform the following operations:

receiving a second mapping of the one or more interrupt channels to the one or more processors; and

distributing the one or more interrupts from the particular device among the one or more processors based on the second mapping.

25. The machine-readable medium of claim 24, wherein the second mapping is received in response to determining that the load from the particular device has exceed a threshold load for the one or more processors, and wherein the second mapping causes the processor load caused by the particular device to fall below the threshold.

26. The machine-readable medium of claim 25, wherein the set of instructions causes the one or more processors to further perform the following operations:

determining that a particular device does not use enough interrupt channels to comply with the second mapping;

designating a particular interrupt channel of the second device as a managed interrupt channel;

selecting a new processor of the plurality of processors as the managing processor; and

distributing interrupts sent over the managed interrupt channel to the newly selected managing processor, wherein the newly selected managing processor assigns one or more tasks associated with interrupts from the particular device to a set of one or more additional processors from the plurality of processors.