Method and structure for skewed block-cyclic distribution of lower-dimensional data arrays in higher-dimensional processor grids

Title: Method and structure for skewed block-cyclic distribution of lower-dimensional data arrays in higher-dimensional processor grids

Full Record
Other Related Research

Abstract

A method and structure of distributing elements of an array of data in a computer memory to a specific processor of a multi-dimensional mesh of parallel processors includes designating a distribution of elements of at least a portion of the array to be executed by specific processors in the multi-dimensional mesh of parallel processors. The pattern of the designating includes a cyclical repetitive pattern of the parallel processor mesh, as modified to have a skew in at least one dimension so that both a row of data in the array and a column of data in the array map to respective contiguous groupings of the processors such that a dimension of the contiguous groupings is greater than one.

Inventors:

Chatterjee, Siddhartha ^[1]; Gunnels, John A ^[2]

Yorktown Heights, NY
Brewster, NY

Issue Date:: 2011-11-08

Research Org.:: International Business Machines Corporation (Armonk, NY)

Sponsoring Org.:: USDOE

OSTI Identifier:: 1033602

Patent Number(s):: 8,055,878

Application Number:: 11/052,216

Assignee:: International Business Machines Corporation (Armonk, NY)

DOE Contract Number:: B517552

Resource Type:: Patent

Country of Publication:: United States

Language:: English

Subject:: 97 MATHEMATICS AND COMPUTING

Citation Formats


                    Chatterjee, Siddhartha, and Gunnels, John A. Method and structure for skewed block-cyclic distribution of lower-dimensional data arrays in higher-dimensional processor grids.  United States: N. p., 2011. 
        Web.

Copy to clipboard


                    Chatterjee, Siddhartha, & Gunnels, John A. Method and structure for skewed block-cyclic distribution of lower-dimensional data arrays in higher-dimensional processor grids.  United States.

Copy to clipboard


                    Chatterjee, Siddhartha, and Gunnels, John A. Tue .  
        "Method and structure for skewed block-cyclic distribution of lower-dimensional data arrays in higher-dimensional processor grids".  United States.  https://www.osti.gov/servlets/purl/1033602.

Copy to clipboard


                    
@article{osti_1033602,

  title        = {Method and structure for skewed block-cyclic distribution of lower-dimensional data arrays in higher-dimensional processor grids},

  author       = {Chatterjee, Siddhartha and Gunnels, John A},

  abstractNote = {A method and structure of distributing elements of an array of data in a computer memory to a specific processor of a multi-dimensional mesh of parallel processors includes designating a distribution of elements of at least a portion of the array to be executed by specific processors in the multi-dimensional mesh of parallel processors. The pattern of the designating includes a cyclical repetitive pattern of the parallel processor mesh, as modified to have a skew in at least one dimension so that both a row of data in the array and a column of data in the array map to respective contiguous groupings of the processors such that a dimension of the contiguous groupings is greater than one.},

  doi          = {},

  journal      = {},
number       = ,

  volume       = ,

  place        = {United States},

  year         = {2011},

  month        = {11}

}

Copy to clipboard

Patent:

View Patent

Save / Share:

Export Metadata

Save to My Library

Similar Records in DOE Patents and OSTI.GOV collections:

Push plate, mounting assembly, circuit board, and method of assembling thereof for ball grid array packages

Patent Vaughn, Mark R.; Montague, Stephen

A push plate that includes springs in the form of cantilever flexures and an inspection window is disclosed. The push plate provides a known, uniform, down force and minimal torque to a package to be tested. The cantilevers have a known, calculable down force producing stiffness. The window provides for viewing of the package during testing.
Full Text Available
Distributed processor allocation for launching applications in a massively connected processors complex

Patent Pedretti, Kevin [Goleta, CA]

A compute processor allocator architecture for allocating compute processors to run applications in a multiple processor computing apparatus is distributed among a subset of processors within the computing apparatus. Each processor of the subset includes a compute processor allocator. The compute processor allocators can share a common database of information pertinent to compute processor allocation. A communication path permits retrieval of information from the database independently of the compute processor allocators.
Full Text Available
Apparatus and method for heterodyne-generated two-dimensional detector array using a single element detector

Patent Strauss, Charlie E [Santa Fe, NM]

Apparatus and method for heterodyne-generated, two-dimensional detector array using a single detector. Synthetic-array heterodyne detection, permits a single-element optical detector to behave as though it were divided into an array of separate heterodyne detector elements. A fifteen-element synthetic array has successfully been experimentally realized on a single-element detector, permitting all of the array elements to be read out continuously and in parallel from one electrical connection. A CO.sub.2 laser and a single-element HgCdTe photodiode are employed. A different heterodyne local oscillator frequency is incident upon the spatially resolvable regions of the detector surface. Thus, different regions are mapped to differentmore »« less
Full Text Available
Three dimensional stress vector sensor array and method therefor

Patent Pfeifer, Kent Bryant; Rudnick, Thomas Jeffery

A sensor array is configured based upon capacitive sensor techniques to measure stresses at various positions in a sheet simultaneously and allow a stress map to be obtained in near real-time. The device consists of single capacitive elements applied in a one or two dimensional array to measure the distribution of stresses across a mat surface in real-time as a function of position for manufacturing and test applications. In-plane and normal stresses in rolling bodies such as tires may thus be monitored.
Full Text Available
Autothermal and partial oxidation reformer-based fuel processor, method for improving catalyst function in autothermal and partial oxidation reformer-based processors

Patent Ahmed, Shabbir; Papadias, Dionissios D.; Lee, Sheldon H. D.; ...

The invention provides a fuel processor comprising a linear flow structure having an upstream portion and a downstream portion; a first catalyst supported at the upstream portion; and a second catalyst supported at the downstream portion, wherein the first catalyst is in fluid communication with the second catalyst. Also provided is a method for reforming fuel, the method comprising contacting the fuel to an oxidation catalyst so as to partially oxidize the fuel and generate heat; warming incoming fuel with the heat while simultaneously warming a reforming catalyst with the heat; and reacting the partially oxidized fuel with steam usingmore » the reforming catalyst.« less
Full Text Available

Similar Records