DEVELOPING COLLECTIVE OPERATIONS FOR A PARALLEL COMPUTER

Patent №

US 9,501,265

Granted

2016-11-22

Filed 2012

Owner

INTERNATIONAL BUSINESS MACHINES CORPORATION

Lab

IBM Research

AI components

hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

13710684

View graph

Developing collective operations for a parallel computer that includes compute nodes includes: presenting, by a collective development tool, a graphical user interface (‘GUI’) to a collective developer; receiving, by the collective development tool from the collective developer through the GUI, a selection of one or more collective primitives; receiving, by the collective development tool from the collective developer through the GUI, a specification of a serial order of the collective primitives and a specification of input and output buffers for each collective primitive; and generating, by the collective development tool in dependence upon the selection of collective primitives, the serial order of the collective primitives, and the input and output buffers for each collective primitive, executable code that carries out the collective operation specified by the collective primitives.

AI hardwareG06F 8/34G06F 9/542

Google Patents

AI classification

AI hardware0.99

Knowledge representation0.19

Machine learning0.02

Planning0.00

Natural language0.00

Evolutionary computation0.00

Speech0.00

Vision0.00

Ownership

INTERNATIONAL BUSINESS MACHINES CORPORATION

assignment · 298310297

Assignors

ARCHER, CHARLES J., CAREY, JAMES E., SANDERS, PHILIP J., SMITH, BRIAN E.

On an employer assignment, the assignors are typically the inventors.

From the same owner

METRIC LEARNING PREDICTION OF SIMULATION PARAMETERSUS 12,691,869 · 2026 DETERMINATION AND ALLEVIATION OF ROOT CAUSES FOR EMISSION OF POLLUTANTS FROM VEHICLESUS 12,693,686 · 2026 TRAINING AND USING A SENTIMENT MACHINE LEARNING MODULE TO RECEIVE AS INPUT HAPTIC METRIC VALUES TO DETERMINE A SENTIMENT SCORE FOR TEXT TO PROVIDE TO AN INTERACTIVE PROGRAMUS 12,693,779 · 2026 DOMAIN-SPECIFICITY PREDICTION FOR NATURAL LANGUAGE PROCESSINGUS 12,694,208 · 2026 USER INTERFACE AUTOMATION USING NATURAL LANGUAGEUS 12,694,222 · 2026 NATURAL LANGUAGE DATA GENERATION USING AUTOMATED KNOWLEDGE DISTILLATION TECHNIQUESUS 12,694,236 · 2026 NEURAL NETWORK SYSTEMS FOR ABSTRACT REASONINGUS 12,694,281 · 2026 ACCELERATING INFERENCE OF NEURAL NETWORK MODELS VIA DYNAMIC EARLY EXITSUS 12,694,283 · 2026

Cites (64)

Performing A Scatterv Operation On A Hierarchical Tree Network Optimized For Collective OperationsUS 8,565,089 · 2013 Monitoring Operating Parameters In A Distributed Computing System With Active MessagesUS 8,436,720 · 2013 RUNTIME OPTIMIZATION OF AN APPLICATION EXECUTING ON A PARALLEL COMPUTERUS 8,365,186 · 2013 METHODS AND SYSTEMS FOR LAUNCHING APPLICATIONS INTO EXISTING ISOLATION ENVIRONMENTSUS 8,326,943 · 2012 PERFORMING AN ALLREDUCE OPERATION ON A PLURALITY OF COMPUTE NODES OF A PARALLEL COMPUTERUS 8,161,268 · 2012 PERFORMING AN ALLREDUCE OPERATION USING SHARED MEMORYUS 8,161,480 · 2012 SYSTEMS AND METHODS FOR DETERMINING COMPUTE KERNELS FOR AN APPLICATION IN A PARALLEL-PROCESSING COMPUTER SYSTEMUS 8,136,104 · 2012 METHOD AND A SYSTEM FOR RESPONDING LOCALLY TO REQUESTS FOR FILE METADATA ASSOCIATED WITH FILES STORED REMOTELYUS 8,131,825 · 2012 METHODS AND SYSTEMS FOR LAUNCHING APPLICATIONS INTO EXISTING ISOLATION ENVIRONMENTSUS 8,090,797 · 2012 BROADCASTING A MESSAGE IN A PARALLEL COMPUTERUS 7,991,857 · 2011 MECHANISM TO SUPPORT GENERIC COLLECTIVE COMMUNICATION ACROSS A VARIETY OF PROGRAMMING MODELSUS 7,984,448 · 2011 CROSS-LAYER DESIGN TECHNIQUES FOR INTERFERENCE-AWARE ROUTING CONFIGURATION IN WIRELESS MESH NETWORKSUS 7,936,681 · 2011 DATA PROCESSING ARCHITECTURES FOR PACKET HANDLING WHEREIN BATCHES OF DATA PACKETS OF UNPREDICTABLE SIZE ARE DISTRIBUTED ACROSS PROCESSING ELEMENTS ARRANGED IN A SIMD ARRAY OPERABLE TO PROCESS DIFFERENT RESPECTIVE PACKET PROTOCOLS AT ONCE WHILE EXECUTING A SINGLE COMMON INSTRUCTION STREAMUS 7,856,543 · 2010 PERFORMING PROCESS MIGRATION WITH ALLREDUCE OPERATIONSUS 7,853,639 · 2010 DYNAMIC MULTIPOINT TREE REARRANGEMENTUS 7,808,930 · 2010 SYSTEM AND METHOD FOR AUTOMATIC GENERATION OF A HIERARCHICAL TREE NETWORK AND THE USE OF TWO COMPLEMENTARY LEARNING ALGORITHMS, OPTIMIZED FOR EACH LEAF OF THE HIERARCHICAL TREE NETWORKUS 7,725,329 · 2010 MEMORY CONTROL DEVICEUS 7,707,366 · 2010 CONFIGURING COMPUTE NODES OF A PARALLEL COMPUTER IN AN OPERATIONAL GROUP INTO A PLURALITY OF INDEPENDENT NON-OVERLAPPING COLLECTIVE NETWORKSUS 7,673,011 · 2010 LOCATING HARDWARE FAULTS IN A DATA COMMUNICATIONS NETWORK OF A PARALLEL COMPUTERUS 7,646,721 · 2010 IMPLEMENTING LOCKS IN A DISTRIBUTED PROCESSING SYSTEMUS 7,640,315 · 2009 EXECUTING SCATTER OPERATION TO PARALLEL COMPUTER NODES BY REPEATEDLY BROADCASTING CONTENT OF SEND BUFFER PARTITION CORRESPONDING TO EACH NODE UPON BITWISE OR OPERATIONUS 7,600,095 · 2009 SYSTEM FOR ALLOCATING COMPUTING RESOURCES OF DISTRIBUTED COMPUTER SYSTEM WITH TRANSACTION MANAGERUS 7,590,983 · 2009 CLASS NETWORK ROUTINGUS 7,587,516 · 2009 NOVEL MASSIVELY PARALLEL SUPERCOMPUTERUS 7,555,566 · 2009 FACILITATING INTRA-NODE DATA TRANSFER IN COLLECTIVE COMMUNICATIONSUS 7,539,989 · 2009

and 39 more

Related research

Visualizing MPI Collective CommunicationOpenAlex Efficient and Eventually Consistent Collective OperationsOpenAlex PCCL: Process Group-Aware Scalable and Generic Collective Algorithm SynthesizerOpenAlex LOOP PARALLELIZATION AUTOMATION FOR GRAPHICS PROCESSING UNITSSemantic Scholar Programming Distributed Collective Processes in the eXchange CalculusarXiv

Full text

Google Patents