Parallel join based on distributed system OceanBase
免费下载
1 5 Ï
2017 c 9
uÀÆÆ(g,Æ)
Journal of East China Normal University (Natural Science)
No. 5
Sept. 2017
©©©ÙÙÙ???ÒÒÒ: 1000-5641(2017)05-0001-10
ÄÄÄuuu©©©ÙÙÙªªªXXXÚÚÚ OceanBase ¿¿¿111ëëë
M[, X, §, a¥w, ±b=
(uÀÆ OÅÆ^ó§Æ, þ° 200062)
Á: XA^êâO±9©Ùªêâ¥XÚØäZy, êâ;3ÔnÕá
!:®²¤«ª³. 3ù«¹e, A^I?1 E, join Î, Ò¬Ø;/
)~õäDÑd. ¤±, XÛJp©ÙªXÚ¥ join ΠǤïÄ9:. ©
3©Û©Ùªêâ¥XÚ OceanBase 1 nested loop join!Hash join!semi-join {
Ä:þ, JÑÜn|^M]æ^õ§¿11 join ö`zg, ¿3 OceanBase
ê⥥©Oé nested loop join!Hash join!semi-join {?1¿1UE. ¢(JL
², 3½§êS join {1Ç¿1Ý¥'.
'c: Î; semi-join; OceanBase; ¿1ë
¥ã©aÒ: TP392 ©zIè: A DOI: 10.3969/j.issn.1000-5641.2017.05.001
Parallel join based on distributed system OceanBase
XU Shi-lei, WANG Lei, HU Hui-qi, QIAN Wei-ning, ZHOU Ao-ying
(School of Computer Science and Software Engineering, East China Normal University,
Shanghai 200062, China )
Abstract: With the rapid growth of application data and the continued development
of distributed database sy stems, data storage in physical independent nodes has become
a trend. In this tren d, when the application needs to perform complex join queries, it
inevitably generates a lot of network traffic. Therefore, improving the efficiency of join
query in distributed system is a hot topic. Based on the analysis of the nested loop join,
Hash join, semi-join in the OceanBase, this paper puts forward the optimization idea of
using hardware resources reasonably and using multithread to execute join operations in
parallel. We implement experiment on OceanBase with nested loop join algorithm, Hash
join algorithm, semi-join algorithm respectively. The experimental results confirm that
the efficiency of join algorithm is positively related to parallelism in a certain number of
threads.
Key words: query; semi-join; OceanBase; parallel join
ÂvFÏ: 2017-06-19
Ä78: 2017cþ°½cE=â~Oy(17YF1427800)
1ö: M[, I, a¬ïÄ), ïÄêâ;êâ÷. E-mail: xsl118857@sina.com.
Ï&ö: §, I, ÏnïÄ, ïÄêâ¥. E-mail: hqhu@dase.ecnu.edu.cn.
2 uÀÆÆ(g,Æ) 2017 c
0 Ú ó
OceanBase
[1]
´Cpnnúimu©Ùªêâ¥, ÙO8I´|±êz TB êâ
þ±9ê TPS! êz QPS ¯þ. Ïd, XÛ¢yéXd êâ¯ÎÑ
Ö, ·óJÑã]Ô.
ÏLé©Ùªêâ¥XÚ OceanBase e©Û, ±, , OceanBase kõê
â;!:ÚõÎ?n!:, ´, 8cÎ?n!:3?n join (ë)óÿ
Ã{¢yÓó. Ïd, Ã{ÏLò?Ö©)õÎ!:Óóª¢yp
¯Î. d, 3Î?n!:¥, éu join Îö, OceanBase ´{ü/¢yG1
z1, ¿vkÜn|^M]ÚêâP{;A5, ddÎ 54äD
ÑÚ join m, XÚ3?nLêâ join Î Ç~$e. éù":, ·J
Ñ«=k/¿1ë`zY: òêâ?1©, âêâõ°A5,
õ§Öêâ; z§ÕáÖêâ, êâÕá1 join ö. l¢y
Î!:¥ join ö¿1z1 , 4/Jp OceanBase E, join ÎÇ.
©ïÄ:´ nested loop join!Hash join!semi-join join {¿1zO±9êâ
©¡©ª.
Ø©SN|Xe: 1 1 !{0OceanBase êâ¥N(9DÚ nested
loop join, merge sort join!Hash join join {Ú semi-join! ©Ùª join {; 1 2 !0
OceanBase ¥DÚ join {ÎnÚÄu semi-join DÚ join {În; 1 3 !
03 OceanBase ¥é nested loop join!Hash join!semi-join join {¿1`zO;
1 4 !l¿1ÝéÎÇK!¿1Ýé join {1ÇK,±9¿1ÝéÄ
u semi-join { join {1ÇK 3 ¡?1 ¡¢y; 1 5 !é
©?1o(.
1 'ó
OceanBase ©Ùªêâ¥NeÌ© 4 ¬: ÌÑÖì RootServer(±e
{¡ RS)! êâ ;ÑÖì ChunkServer(±e{¡ CS)! OþêâÑ Öì Update-
Server(±e{¡ UPS)±9Î?nÑÖì MergeServer(±e{¡ MS). RS KI+n8+
¥¤kÑÖì, Ìü RS; UPS ÌKI?nXÚOþêâ#,
Ìü UPS; MS KKIÂÚ)Û^r SQL ¦, ²Lc{©Û! { ©
Û!Î`zXö=uA CS ½ö UPS; CS Ì KI;XÚÄ Oê
â, ÄOêâ;ün°, .
DÚë{ki@Ìë(nested loop join){! 8¿üSë(merge sort
join){±9MFë(Hash join){. ù 3 «{JÑÑkA½¯KµÚ·^¹.
X©Ùªêâ¥uÐÚÊ9, Bernstein
[2]
qJÑ semi-join (ë){, d{
Ì~ join L§¥êâDÑd. d, 3©ÙªXÚ¥, X MapReduce!Spark
XÚuÐ, 5 õ©ÙªXÚ m©æ^aq Map/Reduce
[3]
O., ùaXÚ
3ÏL?Ö©)ª, ~ join Od. e¡{0ù{A:±9ïÄuÐ
¹.
1.1 nested loop join {
Blasgen
[4]
3 1977 cJÑ nested loop join {. T{·ÜuüLêâþ¿
of 10
免费下载
【版权声明】本文为墨天轮用户原创内容，转载时必须标注文档的来源（墨天轮），文档链接，文档作者等基本信息，否则作者和墨天轮有权追究责任。如果您发现墨天轮中有涉嫌抄袭或者侵权的内容，欢迎发送邮件至：contact@modb.pro进行举报，并提供相关证据，一经查实，墨天轮将立刻删除相关内容。
下载排行榜
评论