Method, apparatus and computer program product for queueing I/O requests on mapped RAID
There is disclosed techniques for queuing I/O requests on Mapped RAID. The techniques comprising queuing a pending I/O request in a queue. The techniques also comprising determining that sufficient credits are available to enable a number of storage devices of a plurality of storage devices in a Mapped RAID group to process the pending I/O request. The techniques further comprising processing the pending I/O request upon determining that there is sufficient credits.
1. A method, comprising:
queuing a pending input/output (I/O) request in a queue;
determining that sufficient credits are available to enable a number of storage devices of a plurality of storage devices in a Mapped redundant array of independent disks (RAID) group to process the pending I/O request, wherein the Mapped RAID group comprises one or more RAID extents, each RAID extent containing a respective set of device extents allocated to that RAID extent, each device extent supported by an extent of storage on a storage device of the plurality of storage devices, wherein credits include one or more of: user I/O credits, background I/O credits, and disk I/O credits;
wherein each RAID extent includes an amount of user I/O credits corresponding to each extent's ability to handle the pending I/O request corresponding to one or more of read/write I/Os;
wherein each RAID extent includes an amount of background I/O credits corresponding to each extent's ability to handle the pending I/O request corresponding to one or more of rebuild, verify, and copy I/Os;
wherein each disk extent of the plurality of storage devices includes an amount of disk I/O credits corresponding to each disk's ability to handle the pending I/O request, each disk having an initial amount of disk I/O credits based on disk type; and
processing the pending I/O request upon determining that there are sufficient credits.
2. The method as claimed in claim 1 , wherein the queue comprises a multi-priority queue with respective priority levels in the multi-priority queue describing different levels of service with respect to processing of pending I/O requests.
3. The method as claimed in claim 2 , wherein one of the priority levels in the multi-priority queue describes a higher level of service relative to another priority level in the multi-priority queue such that a pending I/O request in a higher priority level will be prioritized for processing over a pending I/O request in the other priority level.
4. The method as claimed in claim 2 , wherein respective pending I/O requests in a priority level in the multi-priority queue will be selected for processing in accordance with a first-in first-out approach.
5. The method as claimed in claim 2 , wherein each of the priority levels in the multi-priority queue include a count associated therewith relating to an amount of pending I/O requests to be processed before moving to a next priority level in the multi-priority queue.
6. The method as claimed in claim 5 , wherein a count associated with a priority level in the multi-priority queue is higher relative to a count associated with another priority level in the multi-priority queue with a lower level of servicing.
7. An apparatus, comprising:
memory; and
processing circuitry coupled to the memory, the memory storing instructions which, when executed by the processing circuitry, cause the processing circuitry to:
queue a pending input/output (I/O) request in a queue;
determine that sufficient credits are available to enable a number of storage devices of a plurality of storage devices in a Mapped redundant array of independent disks (RAID) group to process the pending I/O request, wherein the Mapped RAID group comprises one or more RAID extents, each RAID extent containing a respective set of device extents allocated to that RAID extent, each device extent supported by an extent of storage on a storage device of the plurality of storage devices, wherein credits include one or more of: user I/O credits, background I/O credits, and disk I/O credits;
wherein each RAID extent includes an amount of user I/O credits corresponding to each extent's ability to handle the pending I/O request corresponding to one or more of read/write I/Os;
wherein each RAID extent includes an amount of background I/O credits corresponding to each extent's ability to handle the pending I/O request corresponding to one or more of rebuild, verify, and copy I/Os;
wherein each disk of the plurality of storage devices includes an amount of disk I/O credits corresponding to each disk's ability to handle the pending I/O request, each disk having an initial amount of disk I/O credits based on disk type; and
process the pending I/O request upon determining that there are sufficient credits.
8. The apparatus as claimed in claim 7 , wherein the queue comprises a multi-priority queue with respective priority levels in the multi-priority queue describing different levels of service with respect to processing of pending I/O requests.
9. The apparatus as claimed in claim 8 , wherein one of the priority levels in the multi-priority queue describes a higher level of service relative to another priority level in the multi-priority queue such that a pending I/O request in a higher priority level will be prioritized for processing over a pending I/O request in the other priority level.
10. The apparatus as claimed in claim 8 , wherein respective pending I/O requests in a priority level in the multi-priority queue will be selected for processing in accordance with a first-in first-out approach.
11. The apparatus as claimed in claim 8 , wherein each of the priority levels in the multi-priority queue include a count associated therewith relating to an amount of pending I/O requests to be processed before moving to a next priority level in the multi-priority queue.
12. The apparatus as claimed in claim 11 , wherein a count associated with a priority level in the multi-priority queue is higher relative to a count associated with another priority level in the multi-priority queue with a lower level of servicing.
13. A computer program product having a non-transitory computer readable medium which stores a set of instructions, the set of instructions, when carried out by processing circuitry, causing the processing circuitry to perform a method of:
queuing a pending input/output (I/O) request in a queue;
determining that sufficient credits are available to enable a number of storage devices of a plurality of storage devices in a Mapped redundant array of independent disks (RAID) group to process the pending I/O request, wherein the Mapped RAID group comprises one or more RAID extents, each RAID extent containing a respective set of device extents allocated to that RAID extent, each device extent supported by an extent of storage on a storage device of the plurality of storage devices, wherein credits include one or more of: user I/O credits, background I/O credits, and disk I/O credits;
wherein each RAID extent includes an amount of user I/O credits corresponding to each extent's ability to handle the pending I/O request corresponding to one or more of read/write I/Os;
wherein each RAID extent includes an amount of background I/O credits corresponding to each extent's ability to handle the pending I/O request corresponding to one or more of rebuild, verify, and copy I/Os;
wherein each disk extent of the plurality of storage devices includes an amount of disk I/O credits corresponding to each disk's ability to handle the pending I/O request, each disk having an initial amount of disk I/O credits based on disk type; and
processing the pending I/O request upon determining that there are sufficient credits.
14. The computer program product as claimed in claim 13 , wherein the queue comprises a multi-priority queue with respective priority levels in the multi-priority queue describing different levels of service with respect to processing of pending I/O requests.
15. The computer program product as claimed in claim 14 , wherein one of the priority levels in the multi-priority queue describes a higher level of service relative to another priority level in the multi-priority queue such that a pending I/O request in a higher priority level will be prioritized for processing over a pending I/O request in the other priority level.
16. The computer program product as claimed in claim 14 , wherein respective pending I/O requests in a priority level in the multi-priority queue will be selected for processing in accordance with a first-in first-out approach.
17. The computer program product as claimed in claim 14 , wherein each of the priority levels in the multi-priority queue include a count associated therewith relating to an amount of pending I/O requests to be processed before moving to a next priority level in the multi-priority queue.
18. The computer program product as claimed in claim 17 , wherein a count associated with a priority level in the multi-priority queue is higher relative to a count associated with another priority level in the multi-priority queue with a lower level of servicing.