Storage system employing MRAM and array of solid state disks with integrated switch

• US 9,037,786 B2
• Filed: 08/19/2013
• Issued: 05/19/2015
• Est. Priority Date: 09/23/2011
• Status: Active Grant

• Current Assignee: Avalanche Technology Incorporated
• Original Assignee: Avalanche Technology Incorporated
• Inventors: Asnaashari, Mehdi, Nemazie, Siamack, Mandapuram, Anilkumar
• Primary Examiner(s): Bradley, Matthew
• Application Number:
  US13/970,536
• Publication Number:
  US 20130339587A1
• Time in Patent Office: 638 Days
• Field of Search: 711/103, 711/154
• US Class Current: 711/103
• CPC Class Codes: G06F 11/1441 Resetting or repowering
• CPC Class Codes: G06F 11/20 using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
• CPC Class Codes: G06F 11/2005 using redundant communication controllers
• CPC Class Codes: G06F 11/201 between storage system components
• CPC Class Codes: G06F 11/2092 Techniques of failing over between control units
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 12/06 Addressing a physical block of locations, e.g. base addressing, module addressing, memory dedication G06F12/08 takes precedence
• CPC Class Codes: G06F 2212/1032 Reliability improvement, data loss prevention, degraded operation etc
• CPC Class Codes: G06F 2212/2022 Flash memory

Abstract

A high-availability storage system includes a first storage system and a second storage system. The first storage system includes a first Central Processing Unit (CPU), a first physically-addressed solid state disk (SSD) and a first non-volatile memory module that is coupled to the first CPU. Similarly, the second storage system includes a second CPU and a second SSD. Upon failure of one of the first or second CPUs, or the storage system with the non-failing CPU continues to be operational and the storage system with the failed CPU is deemed inoperational and the first and second SSDs remain accessible.

Claims

1. A high-availability (HA) storage appliance comprising:

a first storage system including,

a first Central Processing Unit (CPU) subsystem;

one or more physically-addressed solid state disks (SSDs) addressable using physical addresses, the physical addresses being associated with user data, the user data provided by a host and stored in or retrieved from the one or more physically-addressed SSDs in blocks;

a first non-volatile memory module coupled to the first CPU subsystem, the first non-volatile memory module including flash tables used to manage blocks in the one or more physically-addressed SSDs, the flash tables including tables used to map logical to physical blocks, the physical blocks identifying the location of stored user data in the one or more physically-addressed SSDs;

a first Peripheral Component Interconnect Express (PCIe) switch coupled to the first CPU subsystem, the one or more physically-addressed SSDs and a first network interface, the first network interface being in communication with the host;

a second storage system including;

a second CPU subsystem coupled to the first CPU subsystem through the first PCIe switch or a second PCIe switch, the second PCIe switch further coupled to the first CPU subsystem, the second CPU subsystem and the one or more physically-addressed SSDs, the host communicatively coupled to the first CPU subsystem or the second CPU subsystem through a second network interface, the second network interface coupled to the second PCIe switch,

wherein the first PCIe switch is operable to send user data, provided by the host, to the first CPU subsystem or to the second CPU subsystem, and the second PCIe switch is operable to send user data, provided by the host, to the second CPU subsystem or to the first CPU subsystem,

further wherein upon failure of the first CPU subsystem, the HA storage appliance continues to be operational by re-routing the user data through the first PCIe switch to the second CPU subsystem.

2. The HA storage appliance of claim 1 wherein upon failure of the first CPU subsystem or the second CPU subsystem, the HA storage appliance continues to be operational using the non-failing one of the first CPU subsystem or the second CPU subsystem.

3. The HA storage appliance of claim 1, wherein at least one of the one or more physically-addressed SSDs has a first and second PCIe ports.

4. The HA storage appliance of claim 1, wherein the first storage system and the second storage system are active.

5. The HA storage appliance of claim 3, wherein the first and the second PCIe ports are active.

6. The HA storage appliance of claim 1, wherein upon failure of the first or the second CPU subsystems or the first or the second PCIe switches, at least one of the one or more physically-addressed SSDs remains accessible to the host.

7. The HA storage appliance of claim 1, wherein upon failure of the first or the second PCIe switch, the non-failing one of the first and second PCIe switches is utilized to access at least one of the one or more physically-addressed SSDs.

8. The HA storage appliance of claim 1, wherein the first storage system includes a first network interface controller (NIC) coupled between the first network interface and the first PCIe switch, and the second storage system includes a second NIC, the second NIC being coupled between the second network interface and the second PCIe switch, the first PCIe switch causing coupling of user data, received by the first NIC, to the first CPU subsystem or causing coupling of the user data through the first NIC to the second CPU subsystem, and the second PCIe switch causing coupling of user data, received by the second NIC, to the second CPU subsystem, or causing coupling of user data, received by the second NIC, to the first CPU subsystem.

9. The HA storage appliance of claim 1, wherein the first NIC and the second NIC are coupled to the host.

10. The HA storage appliance of claim 1, wherein upon failure of the first NIC or the second NIC, the non-failing NIC is used to access at least one of the one or more physically-addressed SSDs.

11. The HA storage appliance of claim 1, wherein upon failure of the first PCIe switch or the second PCIe switch, the first CPU subsystem or the second CPU subsystem configures the HA storage appliance to use the non-failing PCIe switch to access at least one of the one or more physically-addressed SSDs.

12. The HA storage appliance of claim 2, wherein the first storage system further includes more than one NIC and the second storage system includes more than one NIC.

13. The HA storage appliance of claim 1, wherein the first storage system further includes more than one PCIe switch and the second storage system further includes more than one PCIe switch.

14. The HA storage appliance of claim 1, wherein upon failure of one of the first or second PCIe switches, user data, from the host, is re-routed to another PCIe switch.

15. The HA storage appliance of claim 1, wherein the second storage system further includes a second non-volatile memory module coupled to the second CPU subsystem.

16. The HA storage appliance of claim 1, wherein the first CPU subsystem or the second CPU subsystem is operable to monitors the operation of the other one of the first and second CPU subsystems.

17. The HA storage system of claim 1, wherein the first CPU subsystem includes a first PCIe root complex and the second CPU subsystem includes a second PCIe root complex.

18. The HA storage appliance of claim 17, wherein the first PCIe switch is a part of the first PCIe root complex and the second PCIe switch is a part of the second PCIe root complex.

19. The HA storage appliance of claim 1, wherein the second storage system further includes a second non-volatile memory module coupled to the second CPU subsystem, the second non-volatile memory module including flash tables used to manage blocks in the one or more physically-addressed SSD, the flash tables including tables used to map logical to physical blocks, the physical blocks used to identify the location of stored user data in the one or more physically-addressed SSD.

20. The HA storage appliance of claim 19, wherein the flash tables of the first non-volatile memory module are duplicated in the second non-volatile memory module and the flash tables of the second non-volatile memory module are duplicated in the first non-volatile memory module.

21. The HA storage appliance of claim 3, wherein the first PCIe switch is coupled to the second PCIe port of at least one of the one or more physically-addressed SSDs and the first PCIe switch is coupled to the second PCIe port of at least one of the one or more physically-addressed SSDs.

22. A high-availability (HA) storage appliance comprising:

one or more physically-addressed solid state disks (SSDs) addressable using physical addresses associated with user data, the user data provided by a host and to be stored in or retrieved from the one or more physically-addressed SSDs in blocks, each of the one or more physically-addressed SSDs having a first and a second Peripheral Component Interconnect Express (PCIe) ports;

a first apparatus including,

a first Central Processing Unit (CPU) subsystem coupled to the first PCIe port of at least one of the one or more physically-addressed SSDs;

a first non-volatile memory module coupled to the first CPU subsystem, the non-volatile memory module including flash tables used to manage blocks in the one or more physically-addressed SSDs, the flash tables including tables used to map logical to physical blocks, the physical blocks identifying the location of stored user data in the one or more physically-addressed SSDs;

a first interface coupled to the first CPU subsystem, the first interface being in communication with the host;

a second apparatus including,

a second CPU subsystem coupled to the first CPU subsystem, the second CPU subsystem further coupled to the second PCIe port of at least one of the one or more physically-addressed SSDs;

a second non-volatile memory module coupled to the second CPU subsystem, the second non-volatile memory module including flash tables used to manage blocks in the one or more physically-addressed SSDs; and

a second interface coupled to the second CPU subsystem, the second interface being in communication with the host,

wherein upon failure of the first apparatus, the HA storage appliance continues to be operational by re-routing the user data to the second apparatus.

23. The HA storage appliance of claim 22, wherein the first apparatus includes a first PCIe switch coupled between the first CPU subsystem and the first interface and the second apparatus includes a second PCIe switch coupled between the second CPU subsystem and the second interface.

24. The HA storage appliance of claim 22, wherein the first apparatus includes a first NIC coupled between the first CPU subsystem and the first interface and the second apparatus includes a second NIC coupled between the second CPU subsystem and the second interface.

25. The HA storage appliance of claim 24, wherein the first apparatus includes a first PCIe switch coupled between the first CPU subsystem and the first PCIe port of at least one of the one or more physically-addressed SSDs and the second apparatus includes a second PCIe switch coupled between the second CPU subsystem and the second PCIe port of at least one of the one or more physically-addressed SSDs.

26. The HA storage appliance of claim 22, wherein the first and the second PCIe ports are active.

27. The HA storage appliance of claim 22, wherein the first apparatus and the second apparatus are active.

28. The HA storage appliance of claim 22, wherein upon failure of the first CPU subsystem or the second CPU subsystem, the HA storage appliance continues to be operational using a remaining one of the first CPU subsystem or the second CPU subsystem.

29. The HA storage appliance of claim 24, wherein upon failure of the first NIC or the second NIC, the HA storage appliance continues to be operational using a remaining one of the first NIC or the second NIC.

30. The HA storage appliance of claim 22, wherein the flash tables of the first non-volatile memory module and the flash tables of the second non-volatile memory module are duplicated of one another.