Hardware flow classification for data storage services

• US 8,407,413 B1
• Filed: 11/05/2010
• Issued: 03/26/2013
• Est. Priority Date: 11/05/2010
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Yucel, Sakir, Sangpachatanaruk, Chatree
• Primary Examiner(s): Farrokh, Hashem
• Assistant Examiner(s): MACKALL, LARRY T
• Application Number:
  US12/940,958
• Time in Patent Office: 872 Days
• Field of Search: 711/108
• US Class Current: 711/108
• 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 69/22 Parsing or analysis of headers

Abstract

The techniques introduced here provide a system and method for hardware implemented storage service flow classification using content addressable memory. The techniques described here allow for classification and handling of storage service requests according to service level agreement (SLA) or quality of service (QoS) parameters without consuming valuable storage server resources.

Claims

1. A method comprising:

receiving an information element at a network storage server, wherein the information element includes a request for storage services on the network storage server and the information element is sent from a client system;

classifying the information element to a storage flow using information contained in the information element and a hardware-based storage flow classification system that is separate from a processor of the network storage server, the classifying the information element to a storage flow including

identifying a client based on the information contained in the information element,

classifying a storage access protocol for the information element based on the client, and

classifying a target storage entity based on the client and the storage access protocol; and

performing an action on the information element based on the storage flow.

2. The method of claim 1 wherein the hardware based storage flow classification system includes a content addressable memory (CAM).

3. The method of claim 2 wherein the CAM is a ternary CAM (TCAM).

4. The method of claim 1 wherein the action performed is based on a service level agreement for the client.

5. A method comprising:

receiving an information element at a network storage server;

identifying a client based on information in the information element by using a content-addressable memory (CAM);

identifying a storage access protocol for the information element, based on the client by using a ternary content addressable memory (TCAM);

identifying a target storage entity for the information element, based on the client and the storage access protocol by using the TCAM; and

performing an action on the information element based on at least one of the client, the storage access protocol, or the target storage entity and a service level agreement (SLA) for the client.

6. The method of claim 5 wherein the client is identified based on information contained in a network header of the information element.

7. The method of claim 6 wherein when a client cannot be identified based on information contained in a network header, the client is identified using pattern matching on upper layers of the information element.

8. The method of claim 5 wherein the client is determined using label encoded content addressable memory (LECAM).

9. The method of claim 5 wherein identifying a storage access protocol includes using pattern matching on upper layers of the information element.

10. The method of claim 5 wherein identifying using a TCAM includes:

dividing a pattern into one or more sub-patterns according to a width of the TCAM;

loading the one or more sub-patterns into the TCAM;

comparing, in the TCAM, a plurality of portions of an information element payload with the one or more sub-patterns, wherein the plurality of portions of the information element each have a length according to the width of the TCAM;

maintaining a record of sub-pattern matches; and

determining from the record of sub-pattern matches a pattern match.

11. The method of claim 5 wherein the TCAM performs a search based on a result of identifying the client.

12. A network storage server comprising:

an interconnect;

a processor coupled with the interconnect;

a storage adapter coupled with the interconnect and configured to provide access to a plurality of storage devices coupled with the storage server;

a network interface coupled with the interconnect and configured to receive an information element containing information from a client system to be stored on at least on of the plurality of storage devices; and

a hardware implemented storage flow classification system separate from the processor and configured to classify the information element to a storage flow based on the information contained in the information element, the classifying the information element to a storage flow including

identifying a client based on the information contained in the information element,

classifying a storage access protocol for the information element based on the client, and

classifying a target storage entity based on the client and the storage access protocol.

13. The network storage server of claim 12 wherein the processor is configured to perform an action based on a classification of the information element.

14. The network storage server of claim 12 wherein the network interface includes the hardware implemented storage flow classification system.

15. The network storage server of claim 12 wherein the hardware implemented storage flow classification system includes a content addressable memory (CAM).

16. The network storage server of claim 15 wherein the CAM is a ternary CAM (TCAM).

17. A cloud storage network comprising:

a head-end server configured to forward a request for storage services to a storage server, wherein the head-end server includes a hardware-based storage flow classification system configured to classify the request for storage services based on information contained in the request, the classifying the request for storage services including

identifying a client based on the information contained in the request,

classifying a storage access protocol for the storage services based on the client, and

classifying a target storage entity based on the client and the storage access protocol, and wherein the head-end server is further configured to include

a classification tag in the forwarded request; and

a storage server configured to receive the request forwarded from the head-end server, interpret the classification tag included by the head-end server, and to perform an action based on the classification tag.

18. The cloud storage network of claim 17 wherein the head-end storage server further includes a network interface and the network interface includes the hardware-based storage flow classification system.

19. The cloud storage network of claim 17 wherein the hardware-based storage flow classification system includes a content addressable memory (CAM).

20. The cloud storage network of claim 19 wherein the CAM is a ternary CAM (TCAM).