Dynamic and consistent naming of fabric attached storage by a file system on a compute node storing information mapping API system I/O calls for data objects with a globally unique identification

• US 6,148,349 A
• Filed: 02/06/1998
• Issued: 11/14/2000
• Est. Priority Date: 02/06/1998
• Status: Expired due to Term

• Current Assignee: Teradata US, Inc. (Teradata Corporation)
• Original Assignee: NCR Corporation
• Inventors: Muller, P. Keith, Chow, Kit M., Meyer, Michael W., Adamson, Alan P.
• Primary Examiner(s): Lee, Thomas C.
• Assistant Examiner(s): Park, Ilwoo
• Application Number:
  US09/019,933
• Time in Patent Office: 1,012 Days
• Field of Search: 709/214, 710/74, 710/33, 714/44
• US Class Current: 710/33
• CPC Class Codes: G06F 15/17375 One dimensional, e.g. linear array, ring

Abstract

The present invention describes a parallel processing system. The system comprises a plurality of compute nodes for executing applications via a storage application interface having system input/output calls, a plurality of I/O nodes, and a file system implemented in the compute node, for storing information mapping API system input/output calls for the data object with the globally unique identification for the data object. Each I/O node manages a communicatively coupled plurality of storage resources and each has a means for generating a globally unique identification for a data object stored on the storage resource and transmits the globally unique identification and the data object to the compute node via at least one interconnecting fabric providing communication between any of the compute node and any of the I/O nodes.

Claims

1. A method of communicating data in a multi-node computer system comprising a plurality of compute nodes, a plurality of input/output (I/O) nodes communicatively coupled to the compute nodes via at least one interconnecting fabric, each I/O node communicatively coupled to a plurality of storage devices, comprising the steps of:

generating a globally unique identification (ID) for a data extent physically stored in at least one of the plurality of storage devices in the I/O node by reading a globally unique I/O node identifier from an administration node, generating a data extent identifier locally unique to the I/O node, and combining the globally unique I/O node identifier with the locally unique data extent identifier;

binding the globally unique ID to the data extent;

exporting the globally unique ID to the compute nodes via the interconnect fabric; and

creating a local entry point in the compute nodes for the data extent physically stored in the at least one of the plurality of storage devices.

2. The method of claim 1, wherein the I/O node identifier is read from the administration node at I/O node boot up.

3. An apparatus for communicating data in a multi-node computer system comprising a plurality of compute nodes, a plurality of input/output (I/O) nodes communicatively coupled to the compute nodes via at least one interconnecting fabric, each I/O node communicatively coupled to a plurality of storage devices, comprising:

means for generating a globally unique identification (ID) for a data extent physically stored in at least one of the plurality of storage devices in the I/O node by reading a globally unique I/O node identifier from an administration node, generating a data extent identifier locally unique to the I/O node, and combining the globally unique I/O node identifier with the locally unique data extent identifier;

means for binding the globally unique ID to the data extent;

means for exporting the globally unique ID to the compute nodes via the interconnect fabric; and

means for creating a local entry point in the compute nodes for the data extent physically stored in at least one of the plurality of storage devices.

4. The apparatus of claim 3, wherein the I/O node identifier is read from the administration node at I/O node boot up.

10. The apparatus of claim 3, wherein:

the I/O nodes are organized into a plurality of cliques, each clique having a primary I/O node and a secondary I/O node;

the primary I/O node generates and exports the globally unique identification; and

the secondary I/O node exports the globally unique identification when the primary I/O node is inoperative.

5. A program storage device, readable by a computer, tangibly embodying one or more programs of instructions executable by the computer to perform method steps of communicating data in a multi-node computer system comprising a plurality of compute nodes, a plurality of input/output (I/O) nodes communicatively coupled to the compute nodes via at least one interconnecting fabric, each I/O node communicatively coupled to a plurality of storage devices, the method steps comprising the steps of:

generating a globally unique identification (ID) for a data extent physically stored in at least one of the plurality of storage devices in the I/O node by reading a globally unique I/O node identifier from an administration node, generating a data extent identifier locally unique to the I/O node, and combining the globally unique I/O node identifier with the locally unique data extent identifier;

binding the globally unique ID to the data extent;

exporting the globally unique ID to the compute nodes via the interconnect fabric; and

creating a local entry point in the compute nodes for the data extent physically stored in at least one of the plurality of storage devices.

6. The program storage device of claim 5, wherein the I/O node identifier is read from the administration node at I/O node boot up.

7. A parallel processing system, comprising:

a plurality of compute nodes for executing applications via a storage application programming interface (API) having system input/output calls;

a plurality of input/output (I/O) nodes, each managing a communicatively coupled plurality of storage resources and each having a means for generating a globally unique identification for a data object stored on the storage resource and for transmitting the globally unique identification and the data object to the compute node via at least one interconnect fabric providing communication between any of the compute nodes and any of the I/O nodes;

a file system, implemented on the compute node, for storing information mapping API system input/output calls for the data object with the globally unique identification for the data object.

8. The parallel processing system of claim 7, wherein the file system stores information mapping the data object with the I/O node managing the storage resource storing the data object.

9. The parallel processing system of claim 7, wherein the system input/output calls comprise an indirect pointer to the file system designating the globally unique identification.

11. The parallel processing system of claim 7, wherein:

the I/O nodes are organized into a plurality of cliques, each clique having a primary I/O node and a secondary I/O node;

the primary I/O node generates and exports the globally unique identification; and

the secondary I/O node exports the globally unique identification when the primary I/O node is inoperative.