[B! Presto] syuu256ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

syuu256 id:syuu256

Prestoã«é–¢ã™ã‚‹syuu256ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (4)

${{author_name}}$

{{{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}}

ã€ŽPrestoã¨ã¯ä½•ã‹ï¼ŒPrestoã§ä½•ãŒã§ãã‚‹ã‹ã€ - ãƒˆãƒ¬ã‚¸ãƒ£ãƒ¼ãƒ‡ãƒ¼ã‚¿ï¼ˆTreasure Dataï¼‰ãƒ–ãƒã‚°
ãƒˆãƒ¬ã‚¸ãƒ£ãƒ¼ãƒ‡ãƒ¼ã‚¿ã¯ã‚¯ãƒ©ã‚¦ãƒ‰ã§ãƒ‡ãƒ¼ã‚¿ãƒžãƒãƒ¼ã‚¸ãƒ¡ãƒ³ãƒˆã‚µãƒ¼ãƒ“ã‚¹ã‚’æä¾›ã—ã¦ã„ã¾ã™ã€‚ Hadoop Conference Japan 2014 ä»¥å‰ã«å‘ŠçŸ¥ã—ãŸHadoop Conference Japan 2014ã§ï¼Œå¼Šç¤¾Software Architectã®å¤æ©‹ãŒç™ºè¡¨ã—ã¾ã—ãŸã€‚ ãƒ†ãƒ¼ãƒžã¯ï¼ŒFacebookãŒå…¬é–‹ã—ãŸæ–°ã—ã„åˆ†æ•£å‡¦ç†åŸºç›¤ï¼ŒPrestoã€‚å®Ÿã¯FacebookãŒå½¼ã‚‰ã®è¶…å¤§è¦æ¨¡ãªãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆã«å¯¾ã—ã¦ã‚¤ãƒ³ã‚¿ãƒ©ã‚¯ãƒ†ã‚£ãƒ–ã«çµæžœã‚’è¿”ã›ã‚‹ã‚ˆã†ã«ã¨é–‹ç™ºã•ã‚ŒãŸã‚‚ã®ã§ã™ã€‚é–‹ç™ºãŒå§‹ã¾ã£ã¦ã¾ã 2å¹´ã‚‚çµŒã£ã¦ãŠã‚Šã¾ã›ã‚“ãŒï¼Œä»Šã§ã¯ãƒˆãƒ¬ã‚¸ãƒ£ãƒ¼ãƒ‡ãƒ¼ã‚¿ã‚’åˆã‚ã¨ã—ã¦å¤šãã®ãƒãƒƒã‚«ãƒ¼é”ãŒã‚³ãƒŸãƒƒã‚¿ãƒ¼ã¨ã—ã¦å‚åŠ ã™ã‚‹æ´»ç™ºçš„ãªãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã«æˆé•·ã—ã¦ã„ã¾ã™ã€‚ Prestoã¯Hiveã‚„Impalaã¨åŒã˜ã€ŒSQL Query Engineã€ã§ã‚ã‚Šï¼Œç‰¹ã«æ•°ç™¾GBã‚’è¶…ãˆã‚‹å¤§è¦æ¨¡ãƒ‡ãƒ¼ã‚¿ã«å¯¾ã—ã¦ã‚‚ã‚¤ãƒ³ã‚¿ãƒ©ã‚¯ãƒ†ã‚£ãƒ–ãªãƒ¬ã‚¹ãƒãƒ³ã‚¹ã‚’ï¼ˆã‚³ãƒ³ãƒž0ç§’ä»¥ä¸‹ï¼Œé…ãã¦
syuu256 2015/03/23
Presto
ãƒªãƒ³ã‚¯
ã‚¹ãƒžãƒ¼ãƒˆãƒ‹ãƒ¥ãƒ¼ã‚¹ã®ä¸–ç•Œé€²å‡ºã‚’æ”¯ãˆã‚‹ãƒã‚°è§£æžåŸºç›¤ #jawsdays #tech
ã‚¹ãƒžãƒ¼ãƒˆãƒ‹ãƒ¥ãƒ¼ã‚¹ã¯æ˜¨å¹´ã® 10/1 ã«ç±³å›½ç‰ˆã‚’ãƒãƒ¼ãƒ³ãƒã™ã‚‹ã«ã‚ãŸã‚Šã€ãƒã‚°è§£æžåŸºç›¤ã®ãƒªãƒ‹ãƒ¥ãƒ¼ã‚¢ãƒ«ã‚’è¡Œã„ã¾ã—ãŸã€‚æ—¥æœ¬ã«åŠ ãˆã¦ç±³å›½ã‚„ãã®ä»–ã®å›½ãŒå…¥ã£ã¦ãã‚‹ã“ã¨ã«ã‚ˆã‚Šã€å˜ãªã‚‹ãƒ¦ãƒ¼ã‚¶æ•°ã®å¢—åŠ ã«åŠ ãˆã€OS x å›½ x ã‚¿ã‚¤ãƒ ã‚¾ãƒ¼ãƒ³ x å¤šç¨®å¤šæ§˜ãªãƒ¡ãƒˆãƒªã‚¯ã‚¹ã®ã‚ˆã†ãªé›†è¨ˆè»¸ãŒå¢—ãˆã‚‹ã“ã¨ã§ã€ãƒã‚°ã®å‰å‡¦ç†ã€é›†è¨ˆã€å¯è¦–åŒ–ã«æ§˜â€¦
syuu256 2015/03/22
Presto

EMR

hive

smartNews
ãƒªãƒ³ã‚¯
Facebookã€åˆ†æ•£SQLã‚¨ãƒ³ã‚¸ãƒ³ã€ŒPrestoã€å…¬é–‹ã€‚å¤§è¦æ¨¡ãƒ‡ãƒ¼ã‚¿ã‚’MapReduce/Hiveã®10å€åŠ¹çŽ‡ã‚ˆãå‡¦ç†ã™ã‚‹ã¨
Facebookã¯ã€æ•°ãƒšã‚¿ãƒã‚¤ãƒˆç´šã®å¤§è¦æ¨¡ãƒ‡ãƒ¼ã‚¿ã«å¯¾ã—ã¦ã‚‚ã€å¯¾è©±çš„ã«ã‚¢ãƒ‰ãƒ›ãƒƒã‚¯ãªå•ã„åˆã‚ã›ã‚’å¯èƒ½ã«ã™ã‚‹åˆ†æ•£SQLã‚¨ãƒ³ã‚¸ãƒ³ã€ŒPrestoã€ã‚’ã€ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ã§å…¬é–‹ã—ã¾ã—ãŸã€‚ Prestoã¯Facebookç¤¾å†…ã§å¤§è¦æ¨¡ãƒ‡ãƒ¼ã‚¿ã®åˆ†æžã®ãŸã‚ã«é–‹ç™ºã•ã‚Œã€ã™ã§ã«åŒç¤¾ç¤¾å†…ä½¿ã‚ã‚Œã¦ã„ã‚‹ã‚‚ã®ã€‚ Facebookã¯Prestoã‚’é–‹ç™ºã—ãŸèƒŒæ™¯ã¨ã—ã¦ã€å¤§é‡ã®ãƒ‡ãƒ¼ã‚¿ã‚’Hadoop/HDFSãƒ™ãƒ¼ã‚¹ã§ä¿å˜ã—ãŸã‚‚ã®ã®ã€ãƒãƒƒãƒæŒ‡å‘ã®MapReduceã§ã¯ãªãã€ãƒªã‚¢ãƒ«ã‚¿ã‚¤ãƒ æ€§ã«å„ªã‚ŒãŸå‡¦ç†ãŒå¿…è¦ã«ãªã£ãŸãŸã‚ã ã¨ã€æ¬¡ã®ã‚ˆã†ã«èª¬æ˜Žã—ã¦ã„ã¾ã™ã€‚ Facebookâ€™s warehouse data is stored in a few large Hadoop/HDFS-based clusters. Hadoop MapReduce [2] and Hive are designed for large-scale, reliabl
syuu256 2013/11/08
Presto

MapReduce

hadoop

Facebook
ãƒªãƒ³ã‚¯
FacebookãŒã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ã®SQLå¯¾å¿œåˆ†æ•£ã‚¯ã‚¨ãƒªã‚¨ãƒ³ã‚¸ãƒ³Prestoã‚’ç™ºè¡¨ã€ãƒšã‚¿ãƒã‚¤ãƒˆè¦æ¨¡ã®ãƒ‡ãƒ¼ã‚¿å‡¦ç†ã‚‚å¯èƒ½
By Intel Free Press ã‚®ã‚¬ãƒã‚¤ãƒˆã‹ã‚‰ãƒšã‚¿ãƒã‚¤ãƒˆ(100ä¸‡ã‚®ã‚¬ãƒã‚¤ãƒˆ)ã¾ã§ã‚ã‚‰ã‚†ã‚‹ã‚µã‚¤ã‚ºã®ãƒ‡ãƒ¼ã‚¿ã‚’ã‚¤ãƒ³ã‚¿ãƒ©ã‚¯ãƒ†ã‚£ãƒ–ã«è§£æžã—ã¦å¿…è¦ãªãƒ‡ãƒ¼ã‚¿ã‚’ç™ºè¦‹å¯èƒ½ã¨ãªã‚‹ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ã®SQLå¯¾å¿œåˆ†æ•£ã‚¯ã‚¨ãƒªã‚¨ãƒ³ã‚¸ãƒ³ãŒPrestoã§ã™ã€‚Facebookã®ã‚ˆã†ãªä¼æ¥ãŒæŒã¤å¤§é‡ã®å•†æ¥ãƒ‡ãƒ¼ã‚¿ã§ã‚‚é«˜é€Ÿã«å‡¦ç†å¯èƒ½ã¨ãªã‚‹ã‚ˆã†ã«è¨è¨ˆã•ã‚Œã¦ãŠã‚Šã€Prestoã®å‡ºã™ã‚¯ã‚¨ãƒª(ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ç®¡ç†ã‚·ã‚¹ãƒ†ãƒ ã«å¯¾ã™ã‚‹å‡¦ç†è¦æ±‚ã‚’æ–‡å—åˆ—ã¨ã—ã¦è¡¨ã—ãŸã‚‚ã®)ã¯è¤‡æ•°ã‚½ãƒ¼ã‚¹ã®ãƒ‡ãƒ¼ã‚¿ã¨çµ„ã¿åˆã‚ã›ã‚‹ã“ã¨ã‚‚å¯èƒ½ã€‚ã“ã®ã‚·ã‚¹ãƒ†ãƒ ã‚’é–‹ç™ºã—ãŸFacebookã§ã¯ã€ç´„300ãƒšã‚¿ãƒã‚¤ãƒˆã«ã‚‚åŠã¶ä¼æ¥å†…éƒ¨ã®ãƒ‡ãƒ¼ã‚¿ã‚’ä¿ç®¡ã™ã‚‹ãƒ‡ãƒ¼ã‚¿ãƒ»ã‚¦ã‚§ã‚¢ãƒã‚¦ã‚¹ã«ã¦Prestoã‚’ä½¿ç”¨ã—ã¦ãŠã‚Šã€æ¯Žæ—¥1000äººã‚’è¶…ãˆã‚‹Facebookã®å¾“æ¥å“¡ãŸã¡ãŒ1æ—¥å½“ãŸã‚Š3ä¸‡ä»¥ä¸Šã®ã‚¯ã‚¨ãƒªã‚’èµ°ã‚‰ã›ã€1ãƒšã‚¿ãƒã‚¤ãƒˆä»¥ä¸Šã®ãƒ‡ãƒ¼ã‚¿ãŒã‚¹ã‚ãƒ£ãƒ³ã•ã‚Œã¦ã„ã‚‹ã€ã¨ã®ã“ã¨ã§ã™ã€‚ Presto | Distribut
syuu256 2013/11/07
Facebook

Presto

ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹
ãƒªãƒ³ã‚¯
1