Parallel matrix multiplication technique optimized for memory fetches

A matrix multiplication circuit comprises a memory storage device, processing circuitry, a parallel multiply circuit, and buffer circuits. The parallel multiply circuit simultaneously performs a count of multiplies in a parallel multiplication operation. The buffer circuits include prefetch buffer c...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chandrasekaran, Praveen, Sugathan, Shreeja, Rajasekar, Vinoth Kumar
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Chandrasekaran, Praveen
Sugathan, Shreeja
Rajasekar, Vinoth Kumar
description A matrix multiplication circuit comprises a memory storage device, processing circuitry, a parallel multiply circuit, and buffer circuits. The parallel multiply circuit simultaneously performs a count of multiplies in a parallel multiplication operation. The buffer circuits include prefetch buffer circuits each having a storage array dimension corresponding to the count of multiplies in the parallel multiplication operation. The processing circuitry loads a first prefetch buffer circuit with values from the first matrix; fetches a value of the second matrix and, in parallel with the fetch, preload the second prefetch buffer circuit with another value from the first matrix; initiates a parallel multiply of the fetched value of the second matrix and the values in the first prefetch buffer circuit; and stores partial product results of the parallel multiply, including adding a current partial product result to a previously stored partial product result.
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_US11556337B2</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>US11556337B2</sourcerecordid><originalsourceid>FETCH-epo_espacenet_US11556337B23</originalsourceid><addsrcrecordid>eNqNyj0KAjEQBtA0FqLeYTyAxRpWe2XFUlDrJcRv2YHJj8ksqKe38QBWr3lz011ccSIQCk4LvyhMopyFvVNOkRR-jPycQCkrB_7gQUMqFBBSedMA9SPq0swGJxWrnwuzPnW343mDnHrU7DwitL9fm6Ztd9buD1v7z_kCa3Y0Sw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Parallel matrix multiplication technique optimized for memory fetches</title><source>esp@cenet</source><creator>Chandrasekaran, Praveen ; Sugathan, Shreeja ; Rajasekar, Vinoth Kumar</creator><creatorcontrib>Chandrasekaran, Praveen ; Sugathan, Shreeja ; Rajasekar, Vinoth Kumar</creatorcontrib><description>A matrix multiplication circuit comprises a memory storage device, processing circuitry, a parallel multiply circuit, and buffer circuits. The parallel multiply circuit simultaneously performs a count of multiplies in a parallel multiplication operation. The buffer circuits include prefetch buffer circuits each having a storage array dimension corresponding to the count of multiplies in the parallel multiplication operation. The processing circuitry loads a first prefetch buffer circuit with values from the first matrix; fetches a value of the second matrix and, in parallel with the fetch, preload the second prefetch buffer circuit with another value from the first matrix; initiates a parallel multiply of the fetched value of the second matrix and the values in the first prefetch buffer circuit; and stores partial product results of the parallel multiply, including adding a current partial product result to a previously stored partial product result.</description><language>eng</language><subject>CALCULATING ; COMPUTING ; COUNTING ; ELECTRIC DIGITAL DATA PROCESSING ; PHYSICS</subject><creationdate>2023</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20230117&amp;DB=EPODOC&amp;CC=US&amp;NR=11556337B2$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20230117&amp;DB=EPODOC&amp;CC=US&amp;NR=11556337B2$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>Chandrasekaran, Praveen</creatorcontrib><creatorcontrib>Sugathan, Shreeja</creatorcontrib><creatorcontrib>Rajasekar, Vinoth Kumar</creatorcontrib><title>Parallel matrix multiplication technique optimized for memory fetches</title><description>A matrix multiplication circuit comprises a memory storage device, processing circuitry, a parallel multiply circuit, and buffer circuits. The parallel multiply circuit simultaneously performs a count of multiplies in a parallel multiplication operation. The buffer circuits include prefetch buffer circuits each having a storage array dimension corresponding to the count of multiplies in the parallel multiplication operation. The processing circuitry loads a first prefetch buffer circuit with values from the first matrix; fetches a value of the second matrix and, in parallel with the fetch, preload the second prefetch buffer circuit with another value from the first matrix; initiates a parallel multiply of the fetched value of the second matrix and the values in the first prefetch buffer circuit; and stores partial product results of the parallel multiply, including adding a current partial product result to a previously stored partial product result.</description><subject>CALCULATING</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>ELECTRIC DIGITAL DATA PROCESSING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2023</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNqNyj0KAjEQBtA0FqLeYTyAxRpWe2XFUlDrJcRv2YHJj8ksqKe38QBWr3lz011ccSIQCk4LvyhMopyFvVNOkRR-jPycQCkrB_7gQUMqFBBSedMA9SPq0swGJxWrnwuzPnW343mDnHrU7DwitL9fm6Ztd9buD1v7z_kCa3Y0Sw</recordid><startdate>20230117</startdate><enddate>20230117</enddate><creator>Chandrasekaran, Praveen</creator><creator>Sugathan, Shreeja</creator><creator>Rajasekar, Vinoth Kumar</creator><scope>EVB</scope></search><sort><creationdate>20230117</creationdate><title>Parallel matrix multiplication technique optimized for memory fetches</title><author>Chandrasekaran, Praveen ; Sugathan, Shreeja ; Rajasekar, Vinoth Kumar</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_US11556337B23</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng</language><creationdate>2023</creationdate><topic>CALCULATING</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>ELECTRIC DIGITAL DATA PROCESSING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>Chandrasekaran, Praveen</creatorcontrib><creatorcontrib>Sugathan, Shreeja</creatorcontrib><creatorcontrib>Rajasekar, Vinoth Kumar</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Chandrasekaran, Praveen</au><au>Sugathan, Shreeja</au><au>Rajasekar, Vinoth Kumar</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Parallel matrix multiplication technique optimized for memory fetches</title><date>2023-01-17</date><risdate>2023</risdate><abstract>A matrix multiplication circuit comprises a memory storage device, processing circuitry, a parallel multiply circuit, and buffer circuits. The parallel multiply circuit simultaneously performs a count of multiplies in a parallel multiplication operation. The buffer circuits include prefetch buffer circuits each having a storage array dimension corresponding to the count of multiplies in the parallel multiplication operation. The processing circuitry loads a first prefetch buffer circuit with values from the first matrix; fetches a value of the second matrix and, in parallel with the fetch, preload the second prefetch buffer circuit with another value from the first matrix; initiates a parallel multiply of the fetched value of the second matrix and the values in the first prefetch buffer circuit; and stores partial product results of the parallel multiply, including adding a current partial product result to a previously stored partial product result.</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language eng
recordid cdi_epo_espacenet_US11556337B2
source esp@cenet
subjects CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
title Parallel matrix multiplication technique optimized for memory fetches
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-22T16%3A40%3A49IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=Chandrasekaran,%20Praveen&rft.date=2023-01-17&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EUS11556337B2%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true