[B! distributed] InoHiroã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

InoHiro id:InoHiro

distributedã«é–¢ã™ã‚‹InoHiroã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (10)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

å¯Œå£«é€šç ”ç©¶æ‰€ãŒæ·±å±¤å¦ç¿’ã®é«˜é€ŸåŒ–æŠ€è¡“ã‚’é–‹ç™ºã€GPU64å°ã§é€Ÿåº¦27å€ã€ã€Œä¸–ç•Œæœ€é«˜é€Ÿã€å®Ÿç¾
å¯Œå£«é€šç ”ç©¶æ‰€ã¯ã€ãƒ‡ã‚£ãƒ¼ãƒ—ãƒ©ãƒ¼ãƒ‹ãƒ³ã‚°ï¼ˆæ·±å±¤å¦ç¿’ï¼šå¤šå±¤ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆã«ã‚ˆã‚‹æ©Ÿæ¢°å¦ç¿’ï¼‰ã‚’è¤‡æ•°ã®ã‚µãƒ¼ãƒãƒ¼ã§åŠ¹çŽ‡çš„ã«åˆ†æ•£å‡¦ç†ã•ã›ã€å¦ç¿’ã‚’é«˜é€ŸåŒ–ã™ã‚‹æŠ€è¡“ã‚’é–‹ç™ºã—ãŸã€‚ã“ã‚Œã¾ã§å¦ç¿’ã«1é€±é–“ã‹ã‹ã£ã¦ã„ãŸã¨ã“ã‚ã‚’ã€6æ™‚é–“ã»ã©ã«çŸç¸®ã§ãã‚‹ã€‚ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆã®è¨è¨ˆã¨æ¤œè¨¼ã®ã‚µã‚¤ã‚¯ãƒ«ã‚’é«˜é€Ÿã«å›žã™ã“ã¨ãŒã§ãã€ç ”ç©¶é–‹ç™ºã®åŠ¹çŽ‡ã‚’é«˜ã‚ã‚‰ã‚Œã‚‹ã€‚ ãƒ‡ãƒ¼ã‚¿å®¹é‡ãŒå¤§ããåˆ†æ•£åŒ–ãŒé›£ã—ã„ã¨ã•ã‚Œã‚‹ç”»åƒèªè˜ç”¨ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆã€ŒAlexNetã€ã§å®Ÿé¨“ã—ãŸã¨ã“ã‚ã€GPU64å°ã®ä¸¦åˆ—åŒ–ã§ã€GPU1å°ã¨æ¯”è¼ƒã—ã¦å‡¦ç†é€Ÿåº¦ã‚’27å€ã«é«˜ã‚ã‚‰ã‚ŒãŸã€‚ã€Œæˆ‘ã€…ãŒç¢ºèªã—ã¦ã„ã‚‹AlexNetã®é«˜é€ŸåŒ–ã§ã¯ã€ã“ã‚ŒãŒä¸–ç•Œã§æœ€ã‚‚é€Ÿã„ã€ï¼ˆå¯Œå£«é€šç ”ç©¶æ‰€ ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ã‚·ã‚¹ãƒ†ãƒ ç ”ç©¶æ‰€ ãƒ‡ãƒ¼ã‚¿ã‚·ã‚¹ãƒ†ãƒ ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆ ä¸»ä»»ç ”ç©¶å“¡ã®ç”°åŽŸå¸ç¦æ°ï¼‰ã€‚ å¯Œå£«é€šç ”ç©¶æ‰€ã¯ã€ä»Šå›žã®æŠ€è¡“ã«ã¤ã„ã¦ç‰¹è¨±ã‚’å‡ºé¡˜æ¸ˆã¿ã€‚å¯Œå£«é€šã®AI æŠ€è¡“ã€ŒHuman Centric AI Zinraiã€ã®ä¸€ã¤ã¨ã—ã¦2016å¹´åº¦ä¸
InoHiro 2016/08/09
distributed

deep_learning
ãƒªãƒ³ã‚¯
GitHub - citusdata/citus: Distributed PostgreSQL as an extension
Citus is a PostgreSQL extension that transf orms Postgres into a distributed databaseâ€”so you can achieve high performance at any scale. With Citus, you extend your PostgreSQL database with new superpowers: Distributed tables are sharded across a cluster of PostgreSQL nodes to combine their CPU, memory, storage and I/O capacity. References tables are replicated to all nodes for joins and foreign key
InoHiro 2016/03/29
postgresql

distributed
ãƒªãƒ³ã‚¯
Hadoopã¯ã©ã®ã‚ˆã†ã«å‹•ãã®ã‹ â”€ä¸¦åˆ—ãƒ»åˆ†æ•£ã‚·ã‚¹ãƒ†ãƒ æŠ€è¡“ã‹ã‚‰èªã¿è§£ãHadoopå‡¦ç†ç³»ã®è¨è¨ˆã¨å®Ÿè£… è¨˜äº‹ä¸€è¦§ | gihyo.jp
ç¬¬21å›žSparkã®è¨è¨ˆã¨å®Ÿè£…ï¼»2ï¼½ï½žSparkã«ãŠã‘ã‚‹ãƒ‡ãƒ¼ã‚¿å…±æœ‰ã®ä»•çµ„ã¿ã¨è€éšœå®³æ€§ã®å®Ÿç¾æ–¹æ³• çŒ¿ç”°æµ©è¼”ï¼Œå±±ç”°æµ©ä¹‹ 2016-06-08
InoHiro 2015/04/18
é¢ç™½ãã†ãªé€£è¼‰

Hadoop

distributed

RDBMS
ãƒªãƒ³ã‚¯
ç¬¬1å›žã€€ãªãœã€Hadoopã¯ã©ã®ã‚ˆã†ã«å‹•ãã®ã‹ã€ã‚’å¦ã¶ã®ã‹ | gihyo.jp
ã¯ã˜ã‚ã« ãƒ“ãƒƒã‚°ãƒ‡ãƒ¼ã‚¿è§£æžã®ãŸã‚ã®ã‚·ã‚¹ãƒ†ãƒ åŸºç›¤ã¨ã—ã¦ã€Hadoopã‚’ã¯ã˜ã‚ã¨ã™ã‚‹ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ã®ãƒ‡ãƒ¼ã‚¿å‡¦ç†ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ï¼ˆãƒ‡ãƒ¼ã‚¿å‡¦ç†ç³»ï¼‰ãŒåºƒãåˆ©ç”¨ã•ã‚Œã¤ã¤ã‚ã‚Šã¾ã™ãŒã€å½“è©²ãƒ‡ãƒ¼ã‚¿å‡¦ç†ç³»ã‚’ã™ã§ã«åˆ©ç”¨ã—ã¦ã„ã‚‹ã€ã‚‚ã—ãã¯åˆ©ç”¨ã®æ¤œè¨Žã‚’ã—ã¦ã„ã‚‹èªè€…ã®æ–¹ã€…ã®ä¸ã«ã¯ã€ãŸã¨ãˆã°ä»¥ä¸‹ã®ã‚ˆã†ãªå•é¡Œã‚’æŠ±ãˆã¦ã„ã‚‹æ–¹ãŒå°‘ãªã‹ã‚‰ãšã„ã‚‰ã£ã—ã‚ƒã‚‹ã®ã§ã¯ãªã„ã§ã—ã‚‡ã†ã‹ã€‚ ãƒ‡ãƒ¼ã‚¿å‡¦ç†ç³»ã®ä½¿ã„æ–¹ã¯ãªã‚“ã¨ãªãã‚ã‹ã‚‹ãŒã€ãã®å†…éƒ¨ã‚’ã‚ã¾ã‚Šç†è§£ã§ãã¦ã„ãªã„ã€‚ã¾ãŸã¯ã€å†…éƒ¨ã®å‹•ä½œåŽŸç†ãŒã‚ˆãã‚ã‹ã‚‰ãªã„ã®ã§ã€æœ¬æ ¼çš„ã«ä½¿ã†æ°—ã«ãªã‚‰ãªã„ã€‚ åŒæ§˜ã®ç›®çš„ã‚’é”æˆã™ã‚‹è¤‡æ•°ã®ãƒ‡ãƒ¼ã‚¿å‡¦ç†ç³»ã«ãŠã„ã¦ã€ã©ã‚Œã‚’ä½¿ã£ã¦è‰¯ã„ã‹ãŒã‚ˆãã‚ã‹ã‚‰ãªã„ã€‚ã¾ãŸã¯ã€é©åˆ‡ã«ä½¿ã„åˆ†ã‘ã‚‰ã‚Œã¦ã„ãªã„æ°—ãŒã™ã‚‹ã€‚ãŸã¨ãˆã°ã€ã©ã®ã‚ˆã†ãªå ´åˆã«Hadoopã‚’ç”¨ã„ã¦ã€ã©ã®ã‚ˆã†ãªå ´åˆã«åŒé¡žã®ãƒ‡ãƒ¼ã‚¿å‡¦ç†ç³»ã§ã‚ã‚‹Impalaã‚„Sparkã‚’ç”¨ã„ã‚Œã°è‰¯ã„ã‹ãŒâ€œâ æ˜Žç¢ºã«â â€ã‚ã‹ã‚‰ãªã„ã€‚ ã“ã®ã‚ˆã†ãªå•é¡Œã‚’è§£æ±ºã™ã‚‹ã«ã¯ã€
InoHiro 2015/04/18
Hadoop

distributed

RDBMS
ãƒªãƒ³ã‚¯
åˆ†æ•£éºä¼çš„ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®èª¬æ˜Ž
åˆ†æ•£éºä¼çš„ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ï¼ˆDistributed Genetic Algorithm : DGAï¼‰ åˆ†æ•£éºä¼çš„ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ï¼ˆDistributed Genetic Algorithm : DGAï¼‰ã¨ã¯å˜ä¸€æ¯é›†å›£GAï¼ˆSingle Population Genetic Algorithm : SPGAï¼‰ã®æ¯é›†å›£ã‚’è¤‡æ•°ã®åˆ†å‰²æ¯é›†å›£ï¼ˆå³¶ï¼‰ã«åˆ†å‰²ã—ï¼Œå„å³¶å†…ã§éºä¼çš„æ“ä½œã‚’è¡Œã†ã¨ã„ã†GAã®ä¸¦åˆ—åŒ–ãƒ¢ãƒ‡ãƒ«ã®ï¼‘ã¤ã§ã‚ã‚‹ï¼ŽDGAã§ã¯å„å³¶é–“ã§è§£æŽ¢ç´¢ã®æƒ…å ±ã‚’äº¤æ›ã™ã‚‹ãŸã‚ã«ï¼Œç§»ä½ã¨ã„ã†æ“ä½œã‚’è¡Œã†ï¼Ž DGAã®æµã‚Œã‚’å³ä¸‹å›³ã«ç¤ºã™ï¼ŽSPGAã®æµã‚Œã¨ç•°ãªã‚‹ã®ã¯ç§»ä½ï¼ˆMigrationï¼‰ã¨ã„ã†æ“ä½œãŒåŠ ã‚ã£ãŸç‚¹ã§ã‚ã‚‹ï¼Ž ç§»ä½ã¯ï¼Œæ•°ä¸–ä»£ã«ä¸€åº¦ï¼Œå„å³¶å†…ã§é¸ã°ã‚ŒãŸï¼‘ã¤ã¾ãŸã¯è¤‡æ•°å€‹ã®å€‹ä½“ï¼ˆç§»ä½å€‹ä½“ : Migrantï¼‰ã‚’åˆ¥ã®å³¶ã¨äº¤æ›ã™ã‚‹ã“ã¨ã§å®Ÿç¾ã•ã‚Œã‚‹ï¼Žã“ã®ã¨ãï¼Œå„å³¶å†…ã«ãŠã‘ã‚‹ç§»ä½å€‹ä½“ã®å‰²åˆã‚’ç§»ä½çŽ‡ï¼ˆMigration Rateï¼‰ï¼Œä½•ä¸–
InoHiro 2013/07/30
GA

distributed
ãƒªãƒ³ã‚¯
climateprediction.net - Wikipedia
InoHiro 2013/07/30
distributed

wikipedia
ãƒªãƒ³ã‚¯
Folding@home - Wikipedia
Folding@homeï¼ˆãƒ•ã‚©ãƒ¼ãƒ«ãƒ‡ã‚£ãƒ³ã‚°ãƒ»ã‚¢ãƒƒãƒˆãƒ»ãƒ›ãƒ¼ãƒ [4]ã€FAHã€F@hã¨ã‚‚å‘¼ã°ã‚Œã‚‹ï¼‰ã¯ã€ã‚¿ãƒ³ãƒ‘ã‚¯è³ªã®å‹•çš„ãªãµã‚‹ã¾ã„ã‚’ã‚·ãƒŸãƒ¥ãƒ¬ãƒ¼ãƒˆã™ã‚‹ã“ã¨ã§ã€ç§‘å¦è€…ãŒæ§˜ã€…ãªç–¾æ‚£ã«å¯¾ã™ã‚‹æ–°ã—ã„æ²»ç™‚æ³•ã‚’é–‹ç™ºã§ãã‚‹ã‚ˆã†æ”¯æ´ã™ã‚‹ã“ã¨ã‚’ç›®çš„ã¨ã—ãŸåˆ†æ•£ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ãƒ†ã‚£ãƒ³ã‚°ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã§ã‚ã‚‹ã€‚ ã“ã‚Œã«ã¯ã‚¿ãƒ³ãƒ‘ã‚¯è³ªã®ãƒ•ã‚©ãƒ¼ãƒ«ãƒ‡ã‚£ãƒ³ã‚°ã‚„ã‚¿ãƒ³ãƒ‘ã‚¯è³ªã®å‹•ãã®ãƒ—ãƒã‚»ã‚¹ãŒå«ã¾ã‚Œã¦ãŠã‚Šã€ãƒœãƒ©ãƒ³ãƒ†ã‚£ã‚¢ã®ãƒ‘ã‚½ã‚³ãƒ³ä¸Šã§å®Ÿè¡Œã•ã‚Œã‚‹åˆ†åå‹•åŠ›å¦ã‚·ãƒŸãƒ¥ãƒ¬ãƒ¼ã‚·ãƒ§ãƒ³ã«ä¾å˜ã—ã¦ã„ã‚‹ã€‚Folding@homeã¯ã€2000å¹´10æœˆ1æ—¥ã«ãƒ“ã‚¸ã‚§ã‚¤ãƒ»ãƒ‘ãƒ³ãƒ‡æ•™æŽˆã®æŒ‡æ®ã®ã‚‚ã¨ã€ã‚¹ã‚¿ãƒ³ãƒ•ã‚©ãƒ¼ãƒ‰å¤§å¦ã§ç™ºè¶³ã—ãŸã€‚2019å¹´ä»¥é™ã¯ã€ã‚»ãƒ³ãƒˆãƒ«ã‚¤ã‚¹ãƒ»ãƒ¯ã‚·ãƒ³ãƒˆãƒ³å¤§å¦åŒ»å¦éƒ¨ã«æ‹ ç‚¹ã‚’ç½®ãã€ãƒ‘ãƒ³ãƒ‡æ•™æŽˆã®å…ƒæ•™ãˆåã§ã‚ã‚‹ã‚°ãƒ¬ãƒƒã‚°ãƒ»ãƒœãƒ¼ãƒžãƒ³åšå£«ãŒæŒ‡æ®ã—ã¦ã„ã‚‹ã€‚ ã“ã®ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã¯ã€åˆ†æ•£ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ãƒ†ã‚£ãƒ³ã‚°ã‚„ç§‘å¦ç ”ç©¶ã‚’è¡Œã†ãŸã‚ã«ã€ã‚°ãƒ©ãƒ•ã‚£ãƒƒã‚¯ã‚¹å‡¦ç†è£…ç½®ï¼ˆGPUï¼‰ã€ä¸å¤®å‡¦ç†è£…ç½®ï¼ˆCPUï¼‰ã€
InoHiro 2013/07/30
distributed

wikipedia
ãƒªãƒ³ã‚¯
BOINC clientâ€“server technology - Wikipedia
Visual diagram of BOINC clientâ€“server structureBOINC clientâ€“server techno logy[1] refers to the model under which BOINC works. The BOINC framework consists of two layers which operate under the clientâ€“server architecture. Once the BOINC software is installed in a machine, the server starts sending tasks to the client. The operations are performed client-side and the results are uploaded to the serv
InoHiro 2013/07/30
distributed

wikipedia
ãƒªãƒ³ã‚¯
DBIS - Databases and Information Systems
InoHiro 2011/06/15
distributed

SPARQL

rdf

mapreduce
ãƒªãƒ³ã‚¯
S4: Distributed Stream Computing Platform
We've got your back )Buyer Protection ProgramWhen you buy a domain name at Dan.com, youâ€™re automatically covered by our Buyer Protection Program. Our unique & carefully designed domain ownership transfer process is the best rated service in the market. Buyer Protection ProgramWhen you buy a domain name at Dan.com, youâ€™re automatically covered by our unique Buyer Protection Program. Read more about
InoHiro 2011/04/18
distributed
ãƒªãƒ³ã‚¯
1