A free-extendible and ultralow-power nonvolatile multi-core associative coprocessor based on MRAM with inter-core pipeline scheme for large-scale full-adaptive nearest pattern searching

A nonvolatile multi-core associative coprocessor based on the MRAM is developed with open-end design, which achieves the higher power efficiency maintaining the high circuit density, full adaptivity and high-speed at the same time. This proposed coprocessor is proposed applicable for searching neare...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Japanese Journal of Applied Physics 2020-04, Vol.59 (SG), p.SGGB18
Hauptverfasser: Ma, Yitao, Miura, Sadahiko, Honjo, Hiroaki, Ikeda, Shoji, Endoh, Tetsuo
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue SG
container_start_page SGGB18
container_title Japanese Journal of Applied Physics
container_volume 59
creator Ma, Yitao
Miura, Sadahiko
Honjo, Hiroaki
Ikeda, Shoji
Endoh, Tetsuo
description A nonvolatile multi-core associative coprocessor based on the MRAM is developed with open-end design, which achieves the higher power efficiency maintaining the high circuit density, full adaptivity and high-speed at the same time. This proposed coprocessor is proposed applicable for searching nearest pattern from the large-scale normal/pre-clustered data sets. An inter-core pipeline operation scheme is implemented, which absorbs the delay of in-core time-domain minimum searching to ensure the high-speed and flexibly extends the coprocessor by increasing the core count. A self-optimized power gating scheme is also employed, which minimizes the operation power and further reduces it when the cluster number (K) becomes larger. The prototype chip with 12-core is designed and fabricated under 90 nm CMOS/70 nm perpendicular-MTJ hybrid technology, and the chip operation at 100 MHz is demonstrated by measurement results. The average operation power is only 68 μW @K = 24, and more than 40-time higher power efficiency is achieved comparing to the latest conventional researches.
doi_str_mv 10.35848/1347-4065/ab72d0
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2377704608</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2377704608</sourcerecordid><originalsourceid>FETCH-LOGICAL-c273t-5b0f12bf8fcf0f00f6be0cc389e3a444646fd3679c386ad8380c15355516f58e3</originalsourceid><addsrcrecordid>eNo9UdtKAzEQDaJgvXyAbwGfo8nmstvHUryBIog-L9nsxEbSZE22rX6af2faik_DnDlzzgwHoQtGr7hsRHPNuKiJoEpe666uenqAJv_QIZpQWjEiplV1jE5y_iitkoJN0M8M2wRA4GuE0LvOA9ahxys_Ju3jhgxxAwmHGNbR69GV8bLMHDExFWbO0bgCrwGbOKRooCAJdzpDj2PATy-zJ7xx4wK7MELabw1uAO8C4GwWsARsy4bX6R1INroY2JX3RPd62OkG0AnyiAc9FoWAc-nNwoX3M3Rktc9w_ldP0dvtzev8njw-3z3MZ4_EVDUfieyoZVVnG2sstZRa1QE1hjdT4FoIoYSyPVf1tEBK9w1vqGGSSymZsrIBfoou97rlv89VOaX9iKsUimVb8bquqVC0KSy2Z5kUc05g2yG5pU7fLaPtLqF2G0e7jaPdJ8R_AdHPiRE</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2377704608</pqid></control><display><type>article</type><title>A free-extendible and ultralow-power nonvolatile multi-core associative coprocessor based on MRAM with inter-core pipeline scheme for large-scale full-adaptive nearest pattern searching</title><source>IOP Publishing Journals</source><source>Institute of Physics (IOP) Journals - HEAL-Link</source><creator>Ma, Yitao ; Miura, Sadahiko ; Honjo, Hiroaki ; Ikeda, Shoji ; Endoh, Tetsuo</creator><creatorcontrib>Ma, Yitao ; Miura, Sadahiko ; Honjo, Hiroaki ; Ikeda, Shoji ; Endoh, Tetsuo</creatorcontrib><description>A nonvolatile multi-core associative coprocessor based on the MRAM is developed with open-end design, which achieves the higher power efficiency maintaining the high circuit density, full adaptivity and high-speed at the same time. This proposed coprocessor is proposed applicable for searching nearest pattern from the large-scale normal/pre-clustered data sets. An inter-core pipeline operation scheme is implemented, which absorbs the delay of in-core time-domain minimum searching to ensure the high-speed and flexibly extends the coprocessor by increasing the core count. A self-optimized power gating scheme is also employed, which minimizes the operation power and further reduces it when the cluster number (K) becomes larger. The prototype chip with 12-core is designed and fabricated under 90 nm CMOS/70 nm perpendicular-MTJ hybrid technology, and the chip operation at 100 MHz is demonstrated by measurement results. The average operation power is only 68 μW @K = 24, and more than 40-time higher power efficiency is achieved comparing to the latest conventional researches.</description><identifier>ISSN: 0021-4922</identifier><identifier>EISSN: 1347-4065</identifier><identifier>DOI: 10.35848/1347-4065/ab72d0</identifier><language>eng</language><publisher>Tokyo: Japanese Journal of Applied Physics</publisher><subject>Circuit design ; CMOS ; Coprocessors ; High speed ; Pattern search ; Power efficiency</subject><ispartof>Japanese Journal of Applied Physics, 2020-04, Vol.59 (SG), p.SGGB18</ispartof><rights>Copyright Japanese Journal of Applied Physics Apr 1, 2020</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c273t-5b0f12bf8fcf0f00f6be0cc389e3a444646fd3679c386ad8380c15355516f58e3</citedby><cites>FETCH-LOGICAL-c273t-5b0f12bf8fcf0f00f6be0cc389e3a444646fd3679c386ad8380c15355516f58e3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27923,27924</link.rule.ids></links><search><creatorcontrib>Ma, Yitao</creatorcontrib><creatorcontrib>Miura, Sadahiko</creatorcontrib><creatorcontrib>Honjo, Hiroaki</creatorcontrib><creatorcontrib>Ikeda, Shoji</creatorcontrib><creatorcontrib>Endoh, Tetsuo</creatorcontrib><title>A free-extendible and ultralow-power nonvolatile multi-core associative coprocessor based on MRAM with inter-core pipeline scheme for large-scale full-adaptive nearest pattern searching</title><title>Japanese Journal of Applied Physics</title><description>A nonvolatile multi-core associative coprocessor based on the MRAM is developed with open-end design, which achieves the higher power efficiency maintaining the high circuit density, full adaptivity and high-speed at the same time. This proposed coprocessor is proposed applicable for searching nearest pattern from the large-scale normal/pre-clustered data sets. An inter-core pipeline operation scheme is implemented, which absorbs the delay of in-core time-domain minimum searching to ensure the high-speed and flexibly extends the coprocessor by increasing the core count. A self-optimized power gating scheme is also employed, which minimizes the operation power and further reduces it when the cluster number (K) becomes larger. The prototype chip with 12-core is designed and fabricated under 90 nm CMOS/70 nm perpendicular-MTJ hybrid technology, and the chip operation at 100 MHz is demonstrated by measurement results. The average operation power is only 68 μW @K = 24, and more than 40-time higher power efficiency is achieved comparing to the latest conventional researches.</description><subject>Circuit design</subject><subject>CMOS</subject><subject>Coprocessors</subject><subject>High speed</subject><subject>Pattern search</subject><subject>Power efficiency</subject><issn>0021-4922</issn><issn>1347-4065</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><recordid>eNo9UdtKAzEQDaJgvXyAbwGfo8nmstvHUryBIog-L9nsxEbSZE22rX6af2faik_DnDlzzgwHoQtGr7hsRHPNuKiJoEpe666uenqAJv_QIZpQWjEiplV1jE5y_iitkoJN0M8M2wRA4GuE0LvOA9ahxys_Ju3jhgxxAwmHGNbR69GV8bLMHDExFWbO0bgCrwGbOKRooCAJdzpDj2PATy-zJ7xx4wK7MELabw1uAO8C4GwWsARsy4bX6R1INroY2JX3RPd62OkG0AnyiAc9FoWAc-nNwoX3M3Rktc9w_ldP0dvtzev8njw-3z3MZ4_EVDUfieyoZVVnG2sstZRa1QE1hjdT4FoIoYSyPVf1tEBK9w1vqGGSSymZsrIBfoou97rlv89VOaX9iKsUimVb8bquqVC0KSy2Z5kUc05g2yG5pU7fLaPtLqF2G0e7jaPdJ8R_AdHPiRE</recordid><startdate>20200401</startdate><enddate>20200401</enddate><creator>Ma, Yitao</creator><creator>Miura, Sadahiko</creator><creator>Honjo, Hiroaki</creator><creator>Ikeda, Shoji</creator><creator>Endoh, Tetsuo</creator><general>Japanese Journal of Applied Physics</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7U5</scope><scope>8FD</scope><scope>H8D</scope><scope>L7M</scope></search><sort><creationdate>20200401</creationdate><title>A free-extendible and ultralow-power nonvolatile multi-core associative coprocessor based on MRAM with inter-core pipeline scheme for large-scale full-adaptive nearest pattern searching</title><author>Ma, Yitao ; Miura, Sadahiko ; Honjo, Hiroaki ; Ikeda, Shoji ; Endoh, Tetsuo</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c273t-5b0f12bf8fcf0f00f6be0cc389e3a444646fd3679c386ad8380c15355516f58e3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Circuit design</topic><topic>CMOS</topic><topic>Coprocessors</topic><topic>High speed</topic><topic>Pattern search</topic><topic>Power efficiency</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Ma, Yitao</creatorcontrib><creatorcontrib>Miura, Sadahiko</creatorcontrib><creatorcontrib>Honjo, Hiroaki</creatorcontrib><creatorcontrib>Ikeda, Shoji</creatorcontrib><creatorcontrib>Endoh, Tetsuo</creatorcontrib><collection>CrossRef</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>Technology Research Database</collection><collection>Aerospace Database</collection><collection>Advanced Technologies Database with Aerospace</collection><jtitle>Japanese Journal of Applied Physics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ma, Yitao</au><au>Miura, Sadahiko</au><au>Honjo, Hiroaki</au><au>Ikeda, Shoji</au><au>Endoh, Tetsuo</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A free-extendible and ultralow-power nonvolatile multi-core associative coprocessor based on MRAM with inter-core pipeline scheme for large-scale full-adaptive nearest pattern searching</atitle><jtitle>Japanese Journal of Applied Physics</jtitle><date>2020-04-01</date><risdate>2020</risdate><volume>59</volume><issue>SG</issue><spage>SGGB18</spage><pages>SGGB18-</pages><issn>0021-4922</issn><eissn>1347-4065</eissn><abstract>A nonvolatile multi-core associative coprocessor based on the MRAM is developed with open-end design, which achieves the higher power efficiency maintaining the high circuit density, full adaptivity and high-speed at the same time. This proposed coprocessor is proposed applicable for searching nearest pattern from the large-scale normal/pre-clustered data sets. An inter-core pipeline operation scheme is implemented, which absorbs the delay of in-core time-domain minimum searching to ensure the high-speed and flexibly extends the coprocessor by increasing the core count. A self-optimized power gating scheme is also employed, which minimizes the operation power and further reduces it when the cluster number (K) becomes larger. The prototype chip with 12-core is designed and fabricated under 90 nm CMOS/70 nm perpendicular-MTJ hybrid technology, and the chip operation at 100 MHz is demonstrated by measurement results. The average operation power is only 68 μW @K = 24, and more than 40-time higher power efficiency is achieved comparing to the latest conventional researches.</abstract><cop>Tokyo</cop><pub>Japanese Journal of Applied Physics</pub><doi>10.35848/1347-4065/ab72d0</doi></addata></record>
fulltext fulltext
identifier ISSN: 0021-4922
ispartof Japanese Journal of Applied Physics, 2020-04, Vol.59 (SG), p.SGGB18
issn 0021-4922
1347-4065
language eng
recordid cdi_proquest_journals_2377704608
source IOP Publishing Journals; Institute of Physics (IOP) Journals - HEAL-Link
subjects Circuit design
CMOS
Coprocessors
High speed
Pattern search
Power efficiency
title A free-extendible and ultralow-power nonvolatile multi-core associative coprocessor based on MRAM with inter-core pipeline scheme for large-scale full-adaptive nearest pattern searching
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T11%3A28%3A49IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20free-extendible%20and%20ultralow-power%20nonvolatile%20multi-core%20associative%20coprocessor%20based%20on%20MRAM%20with%20inter-core%20pipeline%20scheme%20for%20large-scale%20full-adaptive%20nearest%20pattern%20searching&rft.jtitle=Japanese%20Journal%20of%20Applied%20Physics&rft.au=Ma,%20Yitao&rft.date=2020-04-01&rft.volume=59&rft.issue=SG&rft.spage=SGGB18&rft.pages=SGGB18-&rft.issn=0021-4922&rft.eissn=1347-4065&rft_id=info:doi/10.35848/1347-4065/ab72d0&rft_dat=%3Cproquest_cross%3E2377704608%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2377704608&rft_id=info:pmid/&rfr_iscdi=true