Method and system for queue demultiplexor with size grouping
A method for writing data to submission queues in a storage controller including receiving an input/output (I/O) request from a client application, where the client application is associated with a virtual port and where the virtual port is associated with a physical port. The method further includes determining a size of the I/O request, identifying a queue group based on the size of the I/O request and the virtual port, where the queue group includes submission queues and is associated with the virtual port. The method further includes identifying a submission queue, sending the I/O request to a storage controller over the physical port, where the queue group is located in memory operatively connected to the storage controller and where the storage controller is configured to place the I/O request in the submission queue.
1. A method for writing data to submission queues in a storage controller, comprising:
receiving a first input/output (I/O) request from a client application, wherein the client application is associated with a virtual port and wherein the virtual port is associated with a physical port;
determining a size of the first I/O request;
identifying a queue group based on the size of the first I/O request and the virtual port, wherein the queue group consists of a plurality of submission queues, wherein the queue group is associated with the virtual port;
identifying a submission queue in the plurality of submission queues, wherein the submission queue is not full and wherein identifying the submission queue comprises:
making a first determination that reordering of the first I/O request is permitted;
based on the first determination, making a second determination about which of the plurality of submission queues are non-full to obtain a set of non-full submission queues; and
based on the second determination, selecting one of the set of non-full submission queues, wherein the submission queue is one of the set of non-full submission queues;
sending the first I/O request to a storage controller over the physical port, wherein the queue group is located in memory operatively connected to the storage controller, wherein the storage controller is configured to place the first I/O request in the submission queue.
2. The method of claim 1 , further comprising:
receiving a second I/O request from a second client application, wherein the second client application is associated with a second virtual port;
determining a size of the second I/O request;
identifying a second queue group based on the size of the second I/O request and the second virtual port, wherein the second queue group consists of a second plurality of submission queues, wherein the second queue group is associated with the second virtual port;
identifying a second submission queue in the second plurality of submission queues, wherein the second submission queue is not full;
sending the second I/O request to the storage controller over the physical port, wherein the second queue group is located in the memory operatively connected to the storage controller, wherein the storage controller is configured to place the second I/O request in the second submission queue, and wherein the second virtual port is associated with the physical port.
3. The method of claim 1 , further comprising:
prior to receiving the first I/O request:
receiving, by the client, a virtual port handle associated with the first virtual port, wherein the client application provides the virtual port handle with the first I/O request.
4. The method of claim 1 , wherein determining the size of the first I/O request comprises determining at least one selected from a group consisting of an amount of data to be read to complete the first I/O request and an amount of data to be written to complete the first I/O request.
5. The method of claim 1 , wherein the submission queue is located in a region of the memory that is mapped to a client application address space, wherein the first I/O request is stored in the submission queue using submission queue information, and wherein the submission queue information specifies a memory address in the client application address space corresponding to a physical location in the region of memory.
6. The method of claim 1 , wherein the submission queue is located in a region of the memory, wherein sending the first I/O request to the storage controller comprises directly storing the first I/O request in the region of the memory, wherein the first I/O request is stored in the submission queue using submission queue information, and wherein the submission queue information specifies a memory address in a client application address space corresponding to a physical location in the region of memory.
7. The method of claim 1 , wherein the submission queue is located in a region of the memory, wherein sending the first I/O request to the storage controller comprises directly storing the first I/O request in the region of the memory and updating a hardware register in the storage controller with a new memory address, wherein the new memory address specifies a next available memory address in the submission queue in which to store a subsequent I/O request for the submission queue, and wherein the new memory address is in a client application address space.
8. The method of claim 1 , wherein selecting one of the set of non-full submission queues comprises at least one selected from a group consisting of selecting the submission queue using a round robin algorithm, arbitrarily selecting one of the set of non-full submission queues, and selecting the submission queue as the submission queue in the set of non-full submission queues that includes a least number of queued I/O requests.
9. A method for writing data to submission queues in a storage controller, comprising:
receiving an input/output (I/O) request from a client application, wherein the client application is associated with a virtual port and wherein the virtual port is associated with a physical port;
determining a size of the I/O request;
identifying a queue group based on the size of the I/O request and the virtual port, wherein the queue group consists of a plurality of submission queues, wherein the queue group is associated with the virtual port;
identifying a submission queue in the plurality of submission queues, wherein the submission queue is not full and wherein identifying the submission queue comprises:
making a determination that reordering of the I/O request is not permitted;
based on the determination:
hashing at least one selected from a group consisting of an object ID and an offset ID to obtain a hashed value, wherein the I/O request specifies data associated with the object ID and the offset ID, and
selecting the submission queue using the hashed value; and
sending the I/O request to a storage controller over the physical port, wherein the queue group is located in memory operatively connected to the storage controller, wherein the storage controller is configured to place the I/O request in the submission queue.
10. The method of claim 9 , further comprising:
prior to receiving the I/O request:
receiving, by the client, a virtual port handle associated with the virtual port, wherein the client application provides the virtual port handle with the I/O request.
11. The method of claim 9 , wherein determining the size of the I/O request comprises determining at least one selected from a group consisting of an amount of data to be read to complete the I/O request and an amount of data to be written to complete the I/O request.
12. The method of claim 9 , wherein the submission queue is located in a region of the memory that is mapped to a client application address space, wherein the I/O request is stored in the submission queue using submission queue information, and wherein the submission queue information specifies a memory address in the client application address space corresponding to a physical location in the region of memory.
13. The method of claim 9 , wherein the submission queue is located in a region of the memory, wherein sending the I/O request to the storage controller comprises directly storing the I/O request in the region of the memory, wherein the I/O request is stored in the submission queue using submission queue information, and wherein the submission queue information specifies a memory address in a client application address space corresponding to a physical location in the region of memory.
14. The method of claim 9 , wherein the submission queue is located in a region of the memory, wherein sending the I/O request to the storage controller comprises directly storing the I/O request in the region of the memory and updating a hardware register in the storage controller with a new memory address, wherein the new memory address specifies a next available memory address in the submission queue in which to store a subsequent I/O request for the submission queue, and wherein the new memory address is in a client application address space.