[B! case study] oza_x86ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

oza_x86 id:oza_x86

case studyã«é–¢ã™ã‚‹oza_x86ã®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}}

Hadoopã®æœ€æ–°å‹•å‘ã‚’ã€ŒHadoop World:NY 2009ã€ã®è³‡æ–™ã‹ã‚‰ï¼ˆå¾Œç·¨ï¼‰
Hadoopã¯ã€ã‚°ãƒ¼ã‚°ãƒ«ãŒå¤§è¦æ¨¡åˆ†æ•£ã‚·ã‚¹ãƒ†ãƒ ã®ãŸã‚ã«ç”¨ã„ã¦ã„ã‚‹MapReduceã¨ã„ã†æŠ€è¡“ã‚’ã€ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ã¨ã—ã¦å®Ÿç¾ã™ã‚‹ãŸã‚ã«é–‹ç™ºã•ã‚ŒãŸJavaãƒ™ãƒ¼ã‚¹ã®ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã€‚ã‚¯ãƒ©ã‚¦ãƒ‰å¯¾å¿œã®ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã§ã‚ã‚Šã€æ•°ãƒ†ãƒ©ãƒã‚¤ãƒˆã«ã‚‚ãŠã‚ˆã¶å¤§å®¹é‡ã®ãƒ‡ãƒ¼ã‚¿ã‚’é«˜é€Ÿã‹ã¤ä½Žã‚³ã‚¹ãƒˆã«åˆ†æžã™ã‚‹æ–¹æ³•ã¨ã—ã¦æ³¨ç›®ã‚’é›†ã‚ã¦ã„ã¾ã™ã€‚ å¾Œç·¨ã§ã¯ã€10æœˆ2æ—¥ã«ãƒ‹ãƒ¥ãƒ¼ãƒ¨ãƒ¼ã‚¯ã§é–‹å‚¬ã•ã‚ŒãŸã€ŒHadoop Worldï¼šNY 2009ã€ã®åˆå¾Œã®ã‚»ãƒƒã‚·ãƒ§ãƒ³ã®è³‡æ–™ã«ç›®ã‚’é€šã—ã€èˆˆå‘³æ·±ã„ãƒã‚¤ãƒ³ãƒˆã‚’ç´¹ä»‹ã—ã¾ã—ã‚‡ã†ã€‚åˆå¾Œã¯3ãƒˆãƒ©ãƒƒã‚¯ã«åˆ†ã‹ã‚Œ30ã‚‚ã®ã‚»ãƒƒã‚·ãƒ§ãƒ³ãŒè¡Œã‚ã‚Œã¦ã„ã¾ã—ãŸã€‚ ã“ã®è¨˜äº‹ã¯ã€ŒHadoopã®æœ€æ–°å‹•å‘ã‚’ã€ŒHadoop World:NY 2009ã€ã®è³‡æ–™ã‹ã‚‰ï¼ˆå‰ç·¨ï¼‰ã€ã®ç¶šãã§ã™ã€‚ åˆå¾Œã®ã‚»ãƒƒã‚·ãƒ§ãƒ³è³‡æ–™ã‹ã‚‰ãƒã‚¤ãƒ©ã‚¤ãƒˆã‚’ç´¹ä»‹ ã‚¤ã‚§ãƒ¼ãƒ«å¤§å¦ã®Azza Abouzeidæ°ã¨Kamil Bajda-Pawlikowskiæ°ã¯ã€Hadoopã¨ãƒ‘ãƒ©ãƒ¬ãƒ«
oza_x86 2013/10/30
Hadoop

case study
ãƒªãƒ³ã‚¯
ã‚‚ã—ã‚‚ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã‚µãƒ¼ãƒãŒã‚¯ãƒ©ãƒƒã‚·ãƒ¥ã—ãŸã‚‰
äººã®ä½œã£ãŸã‚‚ã®ã¯å®Œç’§ã§ã¯ãªã„ã€‚å®Œç’§ã§ãªã„ã‚‚ã®ã¯ã‚¯ãƒ©ãƒƒã‚·ãƒ¥ã™ã‚‹ã€‚æ•…ã«ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã¯ã‚¯ãƒ©ãƒƒã‚·ãƒ¥ã™ã‚‹ã€‚ã‚µãƒ¼ãƒãƒãƒ¼ãƒ‰ã‚¦ã‚§ã‚¢ã®æ•…éšœã€OSã®ã‚¯ãƒ©ãƒƒã‚·ãƒ¥ã€ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ãã®ã‚‚ã®ã®ãƒã‚°ãªã©åŽŸå› ã¯æ§˜ã€…ã§ã‚ã‚‹ãŒã€æ°¸é ã«åœæ¢ã™ã‚‹ã“ã¨ãªãå‹•ãç¶šã‘ã‚‹RDBMSã¨ã„ã†ã‚‚ã®ã¯å˜åœ¨ã—ãªã„ã€‚ã‚‚ã¡ã‚ã‚“æ•°å¹´é–“ç„¡åœæ¢ã§å•é¡ŒãŒå‡ºãªã„å ´åˆã‚‚ã‚ã‚‹ãŒã€ã‚¯ãƒ©ãƒƒã‚·ãƒ¥ã™ã‚‹ã¨ãã¯ã™ã‚‹ã®ã§å¯¾ç–ãŒå¿…è¦ã§ã‚ã‚‹ã€‚ã‚‚ã—ã‚‚ã®æ™‚ã«å‚™ãˆã¦æŠœã‹ã‚Šãªã„ã®ãŒã‚¹ãƒžãƒ¼ãƒˆãªã‚ªãƒˆã‚³ã®ã‚¹ã‚¿ã‚¤ãƒ«ã§ã‚ã‚‹ã€‚ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã®ã”åˆ©ç”¨ã¯è¨ˆç”»çš„ã«ã€‚ ã¨æ€ã£ãŸãã“ã®ã‚¢ãƒŠã‚¿ï¼ï¼äººç”Ÿãã‚“ãªã«æ¥½ã§ã¯ã‚ã‚Šã¾ã›ã‚“ã€‚ ã‚‚ã¡ã‚ã‚“RDBMSã¯ãƒˆãƒ©ãƒ³ã‚¶ã‚¯ã‚·ãƒ§ãƒ³ã®Durabilityã‚’ä¿è¨¼ã—ã¦ã„ã‚‹ã®ã§99%ã®å ´åˆã¯ãã‚Œã§OKã ã‘ã©ã€ãã‚Œã¯COMMITãŒæˆåŠŸã—ãŸå ´åˆã®è©±ã€‚COMMITã¯å¤§æŠµã®å ´åˆé«˜é€Ÿãªå‡¦ç†ã§ã‚ã‚‹ãŒã€ãã‚Œã§ã‚‚å‡¦ç†ã«ã‹ã‹ã‚‹æ™‚é–“ã¯ã‚¼ãƒã§ã¯ãªã„ã€‚ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ãŒãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã‚µãƒ¼ãƒã«COMMITã‚’é€ä¿¡ã—ã¦ã‹ã‚‰
oza_x86 2011/11/25
case study

DB
ãƒªãƒ³ã‚¯
ã‚°ãƒªãƒ¼CTOãŒèªžã‚‹ã€å¤§è¦æ¨¡ã‚½ãƒ¼ã‚·ãƒ£ãƒ«ã‚²ãƒ¼ãƒ é–‹ç™ºã®èˆžå°è£
CodeZineç·¨é›†éƒ¨ã§ã¯ã€ç¾å ´ã§æ´»èºã™ã‚‹ãƒ‡ãƒ™ãƒãƒƒãƒ‘ãƒ¼ã‚’ã‚¹ã‚¿ãƒ¼ã«ã™ã‚‹ãŸã‚ã®ã‚«ãƒ³ãƒ•ã‚¡ãƒ¬ãƒ³ã‚¹ã€ŒDevelopers Summitã€ã‚„ã€ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ã®ç”Ÿãã–ã¾ã‚’ãƒ–ãƒ¼ã‚¹ãƒˆã™ã‚‹ãŸã‚ã®ã‚¤ãƒ™ãƒ³ãƒˆã€ŒDevelopers Boostã€ãªã©ã€ã•ã¾ã–ã¾ãªã‚«ãƒ³ãƒ•ã‚¡ãƒ¬ãƒ³ã‚¹ã‚’ä¼ç”»ãƒ»é‹å–¶ã—ã¦ã„ã¾ã™ã€‚
oza_x86 2011/11/22
case study

GREE
ãƒªãƒ³ã‚¯
Redis to the Rescue?
oza_x86 2011/11/21
case study
ãƒªãƒ³ã‚¯
ã‚µãƒ¼ãƒ“ã‚¹ï¼†ã‚½ãƒªãƒ¥ãƒ¼ã‚·ãƒ§ãƒ³ | æ ªå¼ä¼šç¤¾ã‚¯ãƒ¬ã‚¹ã‚³
å½“Webã‚µã‚¤ãƒˆã§ã¯ã€ãŠå®¢æ§˜ã®åˆ©ä¾¿æ€§å‘ä¸Šã‚„ã€ã‚µã‚¤ãƒˆæ”¹å–„ã®ãŸã‚ã«Cookieã‚’ä½¿ç”¨ã—ã¦ã„ã¾ã™ã€‚è©³ç´°ã«ã¤ã„ã¦ã¯ã€Œã‚µã‚¤ãƒˆã®ã”åˆ©ç”¨ã«ã¤ã„ã¦ã€é …ç•ªï¼–ï¼Žã‚’ã”è¦§ãã ã•ã„ã€‚ã¾ãŸã€å½“Webã‚µã‚¤ãƒˆã«ã¦å–å¾—ã—ãŸæƒ…å ±ã«å€‹äººæƒ…å ±ã¯å«ã¾ã‚Œã¾ã›ã‚“ãŒã€ãƒ—ãƒ©ã‚¤ãƒã‚·ãƒ¼ãƒãƒªã‚·ãƒ¼ã«è¨˜è¼‰ã—ãŸåˆ©ç”¨ç›®çš„ã«ãŠã„ã¦ã€å½“Webã‚µã‚¤ãƒˆã«ã¦å–å¾—ã—ãŸæƒ…å ±ã¨å€‹äººæƒ…å ±ã®é–¢é€£ä»˜ã‘ã‚’è¡Œã†å ´åˆãŒã‚ã‚Šã¾ã™ã€‚ ãªãŠã€ãŠå®¢æ§˜ãŒCookieã‚’ä½¿ç”¨ã™ã‚‹ã“ã¨ã«åŒæ„ã•ã‚Œãªã„å ´åˆã€ã€Œã‚µã‚¤ãƒˆã®ã”åˆ©ç”¨ã«ã¤ã„ã¦ã€é …ç•ªï¼–ï¼Žã«è¨˜è¼‰ã®ã€å„ç¤¾ã‚ªãƒ—ãƒˆã‚¢ã‚¦ãƒˆï¼ˆç„¡åŠ¹åŒ–ï¼‰ãƒšãƒ¼ã‚¸ã«ã¦Cookieç„¡åŠ¹åŒ–ãŒå¯èƒ½ã§ã™ã€‚
oza_x86 2011/11/20
MQ

case study
ãƒªãƒ³ã‚¯
CSKè¨¼åˆ¸ã‚µãƒ¼ãƒ“ã‚¹ã€æ±è¨¼æ¬¡ä¸–ä»£ã‚·ã‚¹ãƒ†ãƒ ã¸ã®å¯¾å¿œã«IBM WebSphere MQ Low Latency Messagingã‚’æŽ¡ç”¨
EnterpriseZineï¼ˆã‚¨ãƒ³ã‚¿ãƒ¼ãƒ—ãƒ©ã‚¤ã‚ºã‚¸ãƒ³ï¼‰ç·¨é›†éƒ¨ã§ã¯ã€æƒ…å ±ã‚·ã‚¹ãƒ†ãƒ æ‹…å½“ã€ã‚»ã‚ãƒ¥ãƒªãƒ†ã‚£æ‹…å½“ã®æ–¹ã€…å‘ã‘ã«ã€EnterpriseZine Dayã€Security Online Dayã€DataTechã¨ã„ã†ã€3ã¤ã®ã‚¤ãƒ™ãƒ³ãƒˆã‚’é–‹å‚¬ã—ã¦ãŠã‚Šã¾ã™ã€‚ãã‚Œãžã‚Œç·¨é›†éƒ¨ç‹¬è‡ªã®åˆ‡ã‚Šå£ã§ã€æ¥ç•Œãƒˆãƒ¬ãƒ³ãƒ‰ã‚„æœ€æ–°äº‹ä¾‹ã‚’ç¶²ç¾…ã€‚æœ€æ–°ã®å‹•å‘ã‚’çŸ¥ã‚‹ã“ã¨ãŒã§ãã‚‹å ´ã¨ã—ã¦ã€å¥½è©•ã‚’å¾—ã¦ã„ã¾ã™ã€‚
oza_x86 2011/11/20
MQ

case study
ãƒªãƒ³ã‚¯
http://cdsweb.cern.ch/record/1321873/files/ATL-SOFT-PROC-2011-010.pdf
oza_x86 2011/11/20
MQ

case study
ãƒªãƒ³ã‚¯
AWS å°Žå…¥äº‹ä¾‹: TellApart | AWS
å°å£²æ¥ã¯ã€å–å¼•ä»¶æ•°ãŒå¤šã„ã‚‚ã®ã®ãƒžãƒ¼ã‚¸ãƒ³ãŒå°‘ãªã„ã¨ã„ã†éŽé…·ãªæ¥ç•Œã§ã™ã€‚ãªã‹ã§ã‚‚ã€e ã‚³ãƒžãƒ¼ã‚¹ã¯ç‰¹ã«ç«¶äº‰ãŒæ¿€ã—ããªã£ã¦ã„ã¾ã™ã€‚ã“ã®æ¥ç•Œã§å‚‘å‡ºã™ã‚‹ã«ã¯ã€æ¥å‹™ä¸Šã®åˆ¤æ–ã‚’ã™ã°ã‚„ãé©åˆ‡ã«è¡Œãˆã‚‹ã“ã¨ãŒå¿…è¦ã«ãªã‚Šã¾ã™ã€‚ã“ã®ã‚ˆã†ãªç†ç”±ã‹ã‚‰ã€ç«¶äº‰å„ªä½æ€§ã‚’æ±‚ã‚ã‚‹ã‚ªãƒ³ãƒ©ã‚¤ãƒ³å°å£²æ¥è€…ã¯ã€æ¥å‹™ã‚’æ”¯æ´ã™ã‚‹ TellApart ã®ã‚ˆã†ãªãƒ†ã‚¯ãƒŽãƒã‚¸ãƒ¼ãƒ—ãƒ©ãƒƒãƒˆãƒ•ã‚©ãƒ¼ãƒ ã«æ³¨ç›®ã—ã¦ã„ã¾ã™ã€‚ Amazon ã¯ã€æŠ€è¡“é¢ã®å¤šæ§˜ãªèª²é¡Œã«å¯¾å‡¦ã§ãã‚‹å”¯ä¸€ã®ãƒ—ãƒ©ãƒƒãƒˆãƒ•ã‚©ãƒ¼ãƒ ã§ã™ã€‚ä¼šç¤¾è¨ç«‹å½“åˆã€ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ãƒªãƒ³ã‚°ãƒãƒ¼ãƒ ã¯å°‘äººæ•°ã§ã—ãŸãŒã€ãã®ã‚ˆã†ãªèª²é¡Œã‚’éžå¸¸ã«çŸæ™‚é–“ã§è§£æ±ºã§ãã¾ã—ãŸã€ TellApart ã®ç·åˆãƒžãƒ¼ã‚±ãƒ†ã‚£ãƒ³ã‚°ãƒ„ãƒ¼ãƒ« (é¡§å®¢ãƒ‡ãƒ¼ã‚¿ãƒ—ãƒ©ãƒƒãƒˆãƒ•ã‚©ãƒ¼ãƒ ã€äºˆæ¸¬çš„é¡§å®¢åˆ†æžã€æ¬¡ä¸–ä»£ãƒ‡ã‚£ã‚¹ãƒ—ãƒ¬ã‚¤åºƒå‘Šãƒªã‚¿ãƒ¼ã‚²ãƒ†ã‚£ãƒ³ã‚°ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³) ã¯ã€ã‚ªãƒ³ãƒ©ã‚¤ãƒ³å°å£²æ¥è€…ãŒæœ€è‰¯ã®é¡§å®¢ã‚„è¦‹è¾¼ã¿å®¢ã‚’ç‰¹å®šã™ã‚‹ (ã¤ã¾ã‚Šã€Œtell apartã€ã®æ„å‘³) ã®ã«å½¹ç«‹ã¡ã¾ã™ã€‚å°
oza_x86 2011/11/18
which serve tens of thousands of qps to hundreds of millions of end users...

case study
ãƒªãƒ³ã‚¯
Monster World at Amazon EC2
oza_x86 2011/11/18
ec2

case study
ãƒªãƒ³ã‚¯
1,000,000 daily users and no cache
Online games pose a few interesting challenges on their backend: A single user generates one http call every few seconds and the balance between data read and write is close to 50/50 which makes the use of a write through cache or other common scaling approaches less effective. Starting from a rather classic rails application as the traffic grew we gradually changed it in order to meet the require
oza_x86 2011/11/18
statistics

case study
ãƒªãƒ³ã‚¯
1