Processor agnostic data storage in a PCIE based shared storage environment

• US 9,285,995 B2
• Filed: 01/14/2015
• Issued: 03/15/2016
• Est. Priority Date: 11/15/2011
• Status: Active Grant

• Current Assignee: Pavilion Data Systems, Inc.
• Original Assignee: Pavilion Data Systems, Inc.
• Inventors: Malwankar, Kiron Balkrishna
• Primary Examiner(s): Rones, Charles
• Assistant Examiner(s): DOAN, HAN V
• Application Number:
  US14/597,094
• Publication Number:
  US 20150127895A1
• Time in Patent Office: 426 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 12/10 Address translation
• CPC Class Codes: G06F 13/1668 Details of memory controller
• CPC Class Codes: G06F 13/4022 using switching circuits, e.g. switching matrix, connection or expansion network G06F13/4009 takes precedence
• CPC Class Codes: G06F 2212/65 Details of virtual memory and virtual address translation
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/0661 Format or protocol conversion arrangements
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

A controller device associated with a disk array that comprises a plurality of storage devices and a processor receives a request from a host machine that is at a remote location from the disk array. The controller device determines that the request is a data request or a control request based on metadata of the request. Responsive to determining that the request is a data request the controller device converts a format of the data request from a first format to a second format that is compatible with the plurality of storage devices. The data request is then routed in the second format directly to at least one storage device of the plurality of storage devices.

Claims

1. A method comprising:

receiving, by a controller device associated with a disk array that comprises a plurality of storage devices and a processor, a request from a host machine that is at a remote location from the disk array, the request having a first format that is a format other than a non-volatile memory express (NVMe) format;

determining, by the controller device, that the request is a data request based on metadata of the request;

responsive to determining that the request is a data request, performing the following comprising;

converting, by the controller device, the data request from the first format to the NVMe format; and

routing the data request in the NVMe format to a storage device of the plurality of storage devices via a switch, bypassing the processor of the disk array, based on a mapping table that identifies locations on the storage device that are accessible to the host machine.

2. The method of claim 1, wherein the first format comprises a format of a network protocol.

3. The method of claim 2, wherein the network protocol comprises one of a fibre channel (FC) protocol or an internet small computer system interface (iSCSI) protocol.

4. The method of claim 1, wherein:

the first format comprises a small computer system interface (SCSI) format and the request comprises a SCSI command;

receiving the request comprises receiving the SCSI command in a SCSI queue; and

converting the data request from the first format to the NVMe format comprises decoding the SCSI command and subsequently converting the SCSI command to an NVMe command.

5. The method of claim 1, wherein converting the data request into the NVMe format comprises creating an NVMe command that corresponds to the data request.

6. The method of claim 5, further comprising:

adding an entry for the NVMe command to a submission queue associated with the storage device;

notifying the storage device of the entry for the NVMe command in the submission queue; and

receiving a posting of an entry for a completion status for the NVMe command at a completion queue associated with the submission queue once the storage device has completed an operation of the NVMe command.

7. The method of claim 6, further comprising:

receiving an interrupt signal from the storage device;

reading the completion status from the completion queue responsive to receiving the interrupt; and

forwarding the completion status to the host machine.

8. The method of claim 1, wherein the request is addressed to a first logical volume presented on the host machine, the method further comprising:

using the mapping table to translate the first logical volume to a second logical volume on the storage device.

9. The method of claim 8, wherein the first logical volume comprises a pool of a plurality of additional virtual volumes, each of the plurality of additional virtual volumes being associated with one of the plurality of storage devices, wherein the second logical volume is one of the plurality of additional virtual volumes in the pool, and wherein the mapping table maps logical addresses of the first logical volume to logical addresses of the plurality of additional virtual volumes.

10. The method of claim 8, further comprising:

translating a logical address of the second logical volume into a physical address of the at least one storage device using a translation table that maps logical addresses of the second logical volume to physical addresses of the storage device.

11. The method of claim 10, wherein the mapping table and the translation table are merged into a single table.

12. The method of claim 1, further comprising:

receiving an additional request from at least one of the host machine or an additional host machine;

determining, by the controller device, that the additional request is a control request based on metadata of the additional request;

directing, by the controller device, the control request to the processor; and

processing the control request by the processor.

13. A device comprising:

a processor;

a controller device; and

a switch to interconnect the processor, the controller device and a plurality of storage devices;

wherein the controller device is to;

receive a request from a remote host machine, the request having a first format that is a format other than a non-volatile memory express (NVMe) format;

determine that the request is a data request based on metadata of the request;

convert the data request from the first format to the NVMe format; and

route the data request in the NVMe format to a storage device of the plurality of storage devices via the switch, bypassing the processor, based on a mapping table that identifies locations on the storage device that are accessible to the remote host machine.

14. The device of claim 13, wherein to convert the data request into the NVMe format the controller device is to create an NVMe command that corresponds to the data request.

15. The device of claim 14, wherein the controller device is further to:

add an entry for the NVMe command to a submission queue associated with the storage device;

notify the storage device of the entry for the NVMe command in the submission queue; and

receive a posting of an entry for a completion status for the NVMe command at a completion queue associated with the submission queue once the storage device has completed an operation of the NVMe command.

16. The device of claim 15, wherein the controller device is further to:

receive an interrupt signal from the storage device;

read the completion status from the completion queue responsive to receiving the interrupt; and

forward the completion status to the remote host machine.

17. The device of claim 13, wherein the request is addressed to a first logical volume presented on the remote host machine, and wherein the controller device is further to:

use the mapping table to translate the first logical volume to a second logical volume on the storage device.

18. The device of claim 17, wherein the first logical volume comprises a pool of a plurality of additional virtual volumes, each of the plurality of additional virtual volumes being associated with one of the plurality of storage devices, wherein the second logical volume is one of the plurality of additional virtual volumes in the pool, and wherein the mapping table maps logical addresses of the first logical volume to logical addresses of the plurality of additional virtual volumes.

19. The device of claim 17, wherein the controller device is further to translate a logical address of the second logical volume into a physical address of the at least one storage device using a translation table that maps logical addresses of the second logical volume to physical addresses of the storage device.

20. The device of claim 13, wherein:

the controller device is further to;

receive an additional request from at least one of the host machine or an additional host machine;

determine that the additional request is a control request based on metadata of the additional request; and

direct the control request to the processor; and

the processor is to process the control request.