Highly-scalable parallel processing computer system architecture

• US 6,247,077 B1
• Filed: 02/06/1998
• Issued: 06/12/2001
• Est. Priority Date: 02/06/1998
• Status: Expired due to Term

• Current Assignee: Teradata US, Inc. (Teradata Corporation)
• Original Assignee: NCR Corporation
• Inventors: Adamson, Alan P., Muller, P. Keith, Chow, Kit M., Meyer, Michael W.
• Primary Examiner(s): Lee, Thomas
• Assistant Examiner(s): Park, Ilwoo
• Application Number:
  US09/020,198
• Time in Patent Office: 1,222 Days
• Field of Search: 714/43, 714/44, 711/112, 711/114, 710/131, 710/74, 709/213-215, 709/239
• US Class Current: 710/74
• CPC Class Codes: G06F 15/17375 One dimensional, e.g. linear array, ring
• CPC Class Codes: H04L 49/101 using crossbar or matrix
• CPC Class Codes: H04L 49/15 Interconnection of switching modules
• CPC Class Codes: H04L 49/201 Multicast operation; Broadcast operation
• CPC Class Codes: H04L 49/30 Peripheral units, e.g. input or output ports

Abstract

A highly-scalable parallel processing computer system architecture is described. The parallel processing system comprises a plurality of compute nodes for executing applications, a plurality of I/O nodes, each communicatively coupled to a plurality of storage resources, and an interconnect fabric providing communication between any of the compute nodes and any of the I/O nodes. The interconnect fabric comprises a network for connecting the compute nodes and the I/O nodes, the network comprising a plurality of switch nodes arranged into more than g(log_bN) switch node stages, wherein b is a total number of switch node input/output ports, and g(x) indicates a ceiling function providing the smallest integer not less than the argument x, the switch node stages thereby providing a plurality of paths between any network input port and network output port.

Claims

1. A parallel processing system, comprising:

a plurality of compute nodes for executing applications;

a plurality of input/output (I/O) nodes, each managing a communicatively coupled plurality of storage resources, and each I/O node projecting an image of storage objects stored on the storage resources to the compute nodes via at least one interconnect fabric providing communication between any of the compute nodes and any of the I/O nodes;

wherein the I/O nodes comprise;

means for generating a globally unique identification for a data object stored on the storage resource, and for binding the globally unique identification to the data object;

means for exporting the globally unique identification to all of the compute nodes via the communication fabric;

means for retrieving the data object from the storage resource in response to a request from the compute node transmitted over the interconnect fabric;

wherein the compute nodes comprise means for creating a local entry point in the compute node for the data object based on the globally unique identification number.

2. The parallel processing system of claim 1, wherein:

each compute node comprises an associated operating system; and

the globally unique identification comprises operating system dependent data enabling use of the data identified by the globally unique identification by the operating system of the compute node.

3. A parallel processing system, comprising:

a plurality of compute nodes for executing applications;

a plurality of input/output (I/O) nodes, each managing a communicatively coupled plurality of storage resources, and each I/O node projecting an image of storage objects stored on the storage resources to the compute nodes via at least one interconnect fabric providing communication between any of the compute nodes and any of the I/O nodes;

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

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.

4. The parallel processing system of claim 3, wherein the secondary I/O) node retrieves the data object from the storage resource in response to a request from the compute node transmitted over the interconnect fabric when the primary I/O node is inoperative.