Virtual non-volatile memory express drive

• US 9,712,619 B2
• Filed: 11/04/2014
• Issued: 07/18/2017
• Est. Priority Date: 11/04/2014
• Status: Active Grant

• Current Assignee: Pavilion Data Systems, Inc.
• Original Assignee: Pavilion Data Systems, Inc.
• Inventors: Malwankar, Kiron Balkrishna, Govindasamy, Raghuraman, Melnic, Dan M.
• Primary Examiner(s): Lai, Michael C
• Application Number:
  US14/532,777
• Publication Number:
  US 20160127468A1
• Time in Patent Office: 987 Days
• CPC Class Codes: G06F 2009/45579 I/O management, e.g. providing access to device drivers or storage
• CPC Class Codes: G06F 2009/45583 Memory management, e.g. access or allocation
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 67/568 Storing data temporarily at an intermediate stage, e.g. caching

Abstract

A processing device receives an input/output (I/O) command generated by a process, the I/O command directed to a virtual storage provided to a host executing the process, wherein the virtual storage comprises a virtual non-volatile memory express (NVMe) drive. The processing device generates a new I/O command based on the received I/O command and encapsulates the new I/O command into a message. The processing device sends the message to at least one of a remote NVMe storage device or a storage server comprising one or more remote NVMe storage devices.

Claims

1. A method comprising:

receiving, by a processing device, an input/output (I/O) command generated by a process, the I/O command directed to a virtual storage provided to a host executing the process, wherein the virtual storage comprises a virtual non-volatile memory express (NVMe) drive that is mapped to a first set of remote NVMe storage devices managed by a first storage server and a second set of NVMe storage devices managed by a second storage server;

generating a first new I/O command for the first storage server based on the received I/O command;

generating a second new I/O command for the second storage server based on the received I/O command;

encapsulating the first new I/O command into a first message;

encapsulating the second new I/O command into a second message;

sending, by the processing device, the first message to the first storage server; and

sending, by the processing device, the second message to the second storage server.

2. The method of claim 1, further comprising:

receiving a notification of a new virtual NVMe drive, the notification identifying an address associated with the new virtual NVMe drive;

sending a discovery request to the address associated with the new virtual NVMe drive; and

connecting to a storage controller of the first storage server, wherein the storage controller maps the new virtual NVMe drive to one or more remote NVMe storage devices of the first set of remote NVMe storage devices.

3. The method of claim 1, further comprising:

receiving a reply message from the first storage server, the reply message encapsulating a response to the first new I/O command;

removing the response from the reply message; and

forwarding the response to the process.

4. The method of claim 1, wherein the first message comprises an Ethernet packet encapsulating the first new I/O command, and wherein the first new I/O command is included in a payload portion of the Ethernet packet.

5. The method of claim 4, wherein the first new I/O command comprises a read command, the method further comprising:

receiving a plurality of additional Ethernet packets from the first storage server;

extracting data from the plurality of additional Ethernet packets; and

providing the data to the process.

6. The method of claim 4, wherein the I/O command comprises a write command to write data to the virtual NVMe drive, the method further comprising:

generating a plurality of additional Ethernet packets;

encapsulating portions of the data in the plurality of additional Ethernet packets; and

sending the plurality of additional Ethernet packets to the first storage server.

7. The method of claim 1, further comprising:

determining that no response to the first new I/O command has been received;

determining whether a threshold amount of time has elapsed since the first message encapsulating the first new I/O command was sent; and

responsive to determining that the threshold amount of time has elapsed, performing the following comprising;

generating a task management command to cancel the first new I/O command;

encapsulating the task management command in a new message; and

sending the new message to the first storage server.

8. A non-transitory computer readable storage medium comprising instructions that, when executed by a processing device of a host, cause the processing device to perform operations comprising:

receiving, by the processing device, an input/output (I/O) command directed to a virtual NVMe drive mounted on the host, wherein the virtual NVMe drive is mapped to a first set of remote NVMe storage devices managed by a first storage server and a second set of NVMe storage devices managed by a second storage server;

generating a first new I/O command for the first storage server based on the I/O command;

generating a second new I/O command for the second storage server based on the I/O command;

encapsulating, by the processing device, the first new I/O command into a first message;

encapsulating the second new I/O command into a second message;

sending, by the processing device, the first message to the first storage server; and

sending, by the processing device, the second message to the second storage server.

9. The non-transitory computer readable storage medium of claim 8, wherein the first message comprises an Ethernet packet encapsulating the first new I/O command, and wherein the first new I/O command is included in a payload portion of the Ethernet packet.

10. The non-transitory computer readable storage medium of claim 9, wherein the first I/O command comprises a read command, the operations further comprising:

receiving a plurality of additional Ethernet packets from the first storage server;

extracting data from the plurality of additional Ethernet packets; and

providing the data to the host.

11. The non-transitory computer readable storage medium of claim 9, wherein the I/O command comprises a write command to write data to the virtual NVMe drive, the operations further comprising:

generating a plurality of additional Ethernet packets;

encapsulating portions of the data in the plurality of additional Ethernet packets; and

sending the plurality of additional Ethernet packets to at least one of the first storage server or the second storage server.

12. The non-transitory computer readable storage medium of claim 8, the operations further comprising:

receiving a notification of a new virtual NVMe drive, the notification identifying an address associated with the new virtual NVMe drive;

sending a discovery request to the address associated with the new virtual NVMe drive; and

connecting to a storage controller of the first storage server, wherein the storage controller maps the new virtual NVMe drive to one or more remote NVMe storage devices of the first set of remote NVMe storage devices.

13. The non-transitory computer readable storage medium of claim 8, the operations further comprising:

receiving a reply message from the first storage server, the reply message encapsulating a response to the first new I/O command;

removing the response from the reply message; and

forwarding the response to a process on the host.

14. The non-transitory computer readable storage medium of claim 8, the operations further comprising:

determining that no response to the first new I/O command has been received;

determining whether a threshold amount of time has elapsed since the first message encapsulating the first new I/O command was sent; and

responsive to determining that the threshold amount of time has elapsed, performing the following comprising;

generating a task management command to cancel the first new I/O command;

encapsulating the task management command in a new message; and

sending the new message to the first storage server.

15. A computing device comprising:

a memory to store instructions for a non-volatile memory express (NVMe) driver; and

a processing device, coupled to the memory, to execute the instructions, wherein the instructions cause the processing device to;

receive an I/O command generated by a process, the I/O command directed to a virtual NVMe drive, wherein the virtual NVMe drive maps to a first set of remote NVMe storage devices managed by a first storage server and a second set of remote NVMe storage devices managed by a second storage server;

generate a first new I/O command for the first storage server based on the received I/O command;

generate a second new I/O command for the second storage server based on the received I/O command;

encapsulate the first new I/O command into a first message;

encapsulate the second new I/O command into a second message;

send the first message to the first storage server; and

send the second message to the second storage server.

16. The computing device of claim 15, wherein the I/O command comprises a read command, and wherein the processing device is further to:

receive a plurality of additional Ethernet packets from at least one of the first storage server or the second storage server;

extract data from the plurality of additional Ethernet packets; and

provide the data to the process.

17. The computing device of claim 15, wherein the I/O command comprises a write command to write data to the virtual NVMe drive, and wherein the processing device is further to:

generate a plurality of additional Ethernet packets;

encapsulate portions of the data in the plurality of additional Ethernet packets; and

send the plurality of additional Ethernet packets to at least one of the first storage server or the second storage server.

18. The computing device of claim 15, wherein the processing device is further to:

determine that no response to the first new I/O command has been received;

determine whether a threshold amount of time has elapsed since the first message encapsulating the first new I/O command was sent; and

responsive to determining that the threshold amount of time has elapsed, perform the following comprising;

generate a task management command to cancel the first new I/O command;

encapsulate the task management command in a new message; and

send the new message to the first storage server.