Automatically segmenting and populating a distributed computing problem

• US 7,467,180 B2
• Filed: 05/29/2003
• Issued: 12/16/2008
• Est. Priority Date: 05/29/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Lehman, Tobin Jon, Kaufman, James H.
• Primary Examiner(s): Vaughn, Jr.; William C.
• Assistant Examiner(s): Mesfin; Yemane
• Application Number:
  US10/448,949
• Publication Number:
  US 20050015571A1
• Time in Patent Office: 2,028 Days
• Field of Search: 709/201, 709/226, 709/248, 718/100, 718/102, 718/106
• US Class Current: 709/201
• CPC Class Codes: G06F 30/23 using finite element methods [FEM] or finite difference methods [FDM]

Abstract

A method is described for automatically segmenting and populating a distributed problem, including determining a bounding space that is at least equal in size to a problem space volume, partitioning the bounding space into a plurality of sub-spaces capable of holding collections of original problem cells, mapping the distributed problem into the partitioned bounding space to populate the sub-spaces with data from the problem, and determining boundaries between the sub-spaces. The method further includes determining coupling interdependencies between sub-spaces, determining complexities associated with boundaries, calculating densities of the data from the problem, and if the densities of two or more contiguous collections fall below a lower level threshold, dynamically joining the two or more contiguous collections, if a particular collection density exceeds an upper threshold, dynamically subdividing the particular collection into a finer granularity, and processing the problem based on the boundaries, the coupling interdependencies, the complexities, and the densities.

Claims

1. A method of automatically segmenting and populating a distributed problem for processing, the method comprising:

determining a bounding space that is at least equal in size to a problem space volume;

partitioning the bounding space into a plurality of sub-spaces capable of holding collections of original problem cells;

mapping the distributed problem into the partitioned bounding space to populate the sub-spaces with data from the problem;

determining boundaries between the sub-spaces;

determining coupling interdependencies between the sub-spaces;

determining complexities associated with the boundaries;

calculating densities of the data from the problem;

if the densities of two or more contiguous collections fall below a lower level threshold, dynamically joining the two or more contiguous collections;

if a particular collection density exceeds an upper threshold, dynamically subdividing the particular collection into a finer granularity; and

processing the problem based on the boundaries, the coupling interdependencies, the complexities, and the densities.