US 9,927,787 B2 · Method and system for managing distributed computing in automation systems

Abstract

Method and a system for managing distributing computing suitable for implementation in an automation system are provided, wherein in a first step, a set of program instructions is partitioned into a plurality of unit blocks, where each unit block comprises at least one program instruction and, in a second set, at least one complementary block corresponding to at least one unit block are identified from the remainder of the plurality of unit blocks, where the complementary blocks are identified based on a comparison between read-write access of global variables in the unit block and corresponding complementary blocks and, in another step, the plurality of unit blocks are executed on a set of multiple processors within the automation system such that at least one complementary block is executed in parallel with a corresponding unit block.

Claims10

Dependent claims

A method for managing distributing computing suitable for use in an automation system, the method comprising:

partitioning a set of program instructions into a plurality of unit blocks, each unit block comprising at least one program instruction;
identifying at least one complementary block corresponding to at least one unit block from a remainder of said plurality of unit blocks, said at least one complementary block being identified based on a comparison between read-write access of global variables in said unit block comprising the at least one program instruction and the at least one corresponding complementary block, and said at least one complementary block from the remainder of said plurality of unit blocks being established such that none of said global variables read-accessed in said remainder of said plurality of unit blocks is write-accessed in said at least one complementary block;
and
executing the plurality of unit blocks on a set of multiple processors within the automation system, the at least one complementary block being executed in parallel with at least one corresponding unit block.

1 more element

A distributed computing management system in an automation system, said distributed computing management system comprising:

a process controller;
an engineering station;
a partitioning module configured to partition a set of program instructions into a plurality of unit blocks, each unit block of the plurality of unit blocks comprising at least one program instruction;
an assessment module configured to identify at least one complementary block corresponding to at least one unit block from a remainder of said plurality of unit blocks, said complementary blocks being identified based on a comparison between read-write access of global variables in said unit block of the plurality of unit blocks and corresponding complementary blocks, and said at least one complementary block from the remainder of said plurality of unit blocks being established such that none of said global variables read-accessed in said remainder of said plurality of unit blocks is write-accessed in said at least one complementary block;
and
a scheduling module configured to execute said plurality of unit blocks on a set of multiple processors within said automation system, at least one complementary block being executed in parallel with a corresponding unit block of the plurality of unit blocks.

3 more elements

Dependencies are inferred from the claim text.

Method and system for managing distributed computing in automation systems

Life of this patent

Method and system for managing distributed computing in automation systems

Life of this patent

Abstract

Claims10