NETWORK STORAGE SYSTEMS HAVING CLUSTERED RAIDS FOR IMPROVED REDUNDANCY AND LOAD BALANCING

• US 20130318297A1
• Filed: 05/24/2012
• Published: 11/28/2013
• Est. Priority Date: 05/24/2012
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Jibbe, Mahmoud K., Sengupta, Dhishankar, Sharma, Abhishekh P., Satchitanand, Karthik Krishnamurthy, Darisa, Venkata Ramprasad
• Granted Patent: US 8,832,372 B2
• Time in Patent Office: Days
• US Class Current: 711/114
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/2005 using redundant communication controllers
• CPC Class Codes: G06F 12/0684 with feedback, e.g. presence or absence of unit detected by addressing, overflow detection
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• 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 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: H04L 45/22 Alternate routing
• CPC Class Codes: H04L 45/30 Routing of multiclass traffic
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• 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]

Abstract

A clustered network-based storage system includes a host server, multiple high availability system controller pairs, and multiple storage devices across multiple arrays. Two independent storage array subsystems each include a quorum drive copy and are each controlled by a HA pair, with remote volume mirroring links coupling the separate HA pairs. The host server includes a virtualization agent that identifies and prioritizes communication paths, and also determines capacity across all system nodes. A system storage management agent determines an overall storage profile across the system. The virtualization agent, storage management agent, quorum drive copies and remote volume minoring link all operate to provide increased redundancy, load sharing, or both between the separate first and second arrays of storage devices.

Claims

1. A clustered network-based storage system, comprising:

a host server adapted to transmit and receive data across the system, said host server including a virtualization agent and a storage management agent, wherein the virtualization agent facilitates identification of clustered system communication paths and determinations of capacity across multiple clustered system nodes, and wherein the storage management agent facilitates the determination of an overall storage profile across the clustered system;

a plurality of storage devices adapted to store data transmitted from the host server, said plurality of storage devices being arranged in separate independent arrays, wherein each array of storage devices includes a quorum drive copy containing information about the overall storage profile of the clustered system; and

a plurality of high availability system controller pairs, each high availability system controller pair including first and second controllers that are each adapted for designation as a clustered system node and for reading and writing to at least some of the plurality of storage devices based upon commands received from the host server, wherein each of the first and second controllers are in communication with each other, the host server, a respectively assigned array of storage devices from the plurality of storage devices, and a corresponding controller in a different high availability controller pair via a remote volume mirroring link, and wherein the virtualization agent, storage management agent, quorum drive copies and remote volume minoring link are adapted to provide increased redundancy, load sharing, or both between the separate independent arrays of storage devices.

2. The clustered network-based storage system of claim 1, wherein said plurality of storage devices includes a plurality of physical disks, and wherein said storage management agent is adapted to facilitate providing information about the overall storage profile of the clustered system to each said quorum drive copy.

3. The clustered network-based storage system of claim 1, wherein the plurality of high availability system controller pairs includes a first controller pair that controls and interacts directly with a first array of storage devices and a second controller pair that controls and interacts directly with a second array of storage devices.

4. The clustered network-based storage system of claim 3, wherein the first controller pair can control the second array of storage devices by operating through its remote volume mirroring links with the second controller pair.

5. The clustered network-based storage system of claim 1, wherein the overall storage profile includes information regarding partition creation, ownership assignment, path priority schema and metadata management.

6. The clustered network-based storage system of claim 1, wherein the storage management agent includes one or more predictive failure analysis modules.

7. The clustered network-based storage system of claim 6, wherein at least one predictive failure analysis module is adapted to facilitate backing up resources from a first array to a second separate array prior to anticipated down time for a clustered system node responsible for the first array.

8. The clustered network-based storage system of claim 1, wherein a total capacity of each array of storage devices is substantially equal to the total capacity of another separate array of storage devices, and wherein said substantially equal capacities facilitates load sharing between the separate arrays of storage devices.

9. The clustered network-based storage system of claim 1, wherein each of said independent arrays of storage devices is arranged such that a first portion of total capacity of the array is reserved for local storage purposes and a second portion of total capacity of the array is reserved for backing up the another separate array of storage devices.

10. The clustered network-based storage system of claim 1, wherein the virtualization agent facilitates the identification of at least four different communication paths to each of the plurality of storage devices from the host server.

11. The clustered network-based storage system of claim 10, wherein the at least four different paths are characterized in a hierarchy that includes at least preferred, primary alternate, secondary alternate, and tertiary alternate paths.

12. A first network-based storage system controller adapted to operate as a high availability paired controller in a clustered network-based storage environment, the system controller comprising:

a first module adapted to facilitate operation of a first system node that controls reading and writing to each of a first plurality of storage devices arranged within a first array of storage devices based upon commands received from a separate host server, said commands including commands from a virtualization agent on the host server that facilitates the identification of clustered system communication paths and facilitates determinations of capacity across multiple clustered system nodes;

a second module adapted to facilitate interactions with a separate second system controller set in a first high availability pair arrangement with said first system controller, wherein the separate second controller operates a second system node that controls the reading and writing to each of a second plurality of storage devices arranged within said first array of storage device based upon commands received from the separate host server; and

a third module adapted to facilitate interactions across a remote volume minoring link with a separate third system controller set in a separate second high availability pair arrangement with a separate fourth system controller, wherein the separate third system controller operates a third system node that controls the reading and writing to each of a third plurality of storage devices arranged within a separate second array of storage devices based upon commands received from the separate host server, and wherein said third module facilitates system redundancies, load sharing, or both between the first and second high availability controller pairs over the remote volume mirroring link.

13. The system controller of claim 12, wherein the commands received from the separate host server further include commands from a storage management agent that facilitates the determination of an overall storage profile across the clustered system

14. The system controller of claim 13, wherein the overall storage profile includes information regarding partition creation, ownership assignment, path priority schema and metadata management.

15. The system controller of claim 13, wherein the storage management agent includes one or more predictive failure analysis modules adapted to facilitate the backing up of resources from the first array to the second separate array prior to anticipated down time for said system controller.

16. The system controller of claim 12, wherein the capacity of the first array of storage devices is substantially equal to the capacity of the second array of storage devices, and wherein said system controller is configured to reserve a first portion of the capacity of the first array of storage devices for local storage purposes and a second portion of its capacity for backing up the second array of storage devices.

17. A method of operating a clustered network-based storage system, the method comprising:

detecting a presence of at least four separate system controllers as individual system nodes coupled to a host server, wherein arrangement of said at least four separate system controllers includes at least two separate high availability controller pairs that control a plurality of storage devices associated, and wherein the arrangement of said plurality of storage devices includes at least a first array of storage devices controlled directly by a first high availability controller pair and a second separate independent array of storage devices controlled directly by a second high availability controller pair;

establishing a plurality of remote volume mirroring links to link controllers between the first and second high availability controller pairs, wherein each remote volume mirroring linked controller is adapted to control indirectly a separate array of storage devices via at least one of the remote volume mirroring links;

exchanging metadata between the individual system nodes, wherein said metadata includes information regarding organization of the clustered network-based storage system;

identifying at least four different communication paths from the host server to each of the storage devices, wherein each of the at least four different communication paths travels through at least one individual system node;

determining an aggregated capacity of the storage devices on the clustered network-based storage system; and

creating an overall storage profile for the clustered network-based storage system, said overall storage profile including controller to storage device ownership properties and communication path priorities, wherein said remote volume mirroring links, exchanged metadata, identified communication paths, determined aggregated capacity and overall storage profile are used to provide increased redundancy, load sharing, or both between the separate first and second arrays of storage devices.

18. The method of claim 17, further comprising:

establishing predictive failure analysis standards with respect to communications across each of the individual system nodes;

receiving a write request from the host server;

determining whether a preferred or a primary alternate path from the host server to a particular storage device from the first array of storage devices is available for the received write request;

determining whether the current communication parameters across the preferred or the primary alternate path meet or exceed the predictive failure analysis standards; and

performing the write request to the particular storage device when the preferred or the primary alternate path is available and when the current communication parameters meet or exceed the predictive failure analysis standards.

19. The method of claim 18, further comprising:

backing up the first array of storage devices onto the second array of storage devices over one or more of the remote volume mirroring links when the current communication parameters fail to meet or exceed the predictive failure analysis standards.

20. The method of claim 17, further comprising:

establishing priority storage standards of storage devices with respect to regular and priority storage of data for the clustered network-based storage system;

receiving a write request from the host server;

determining whether the write request meets the established standards for the priority storage of data;

performing the write request to a priority volume storage device on the first array when the write request is determined to meet established standards for the priority storage of data; and

copying the contents of the priority volume storage device to a backup storage device on the second array over one or more of the remote volume mirroring links.