2019/5/16Ф줿GPU Deep Learning Community #11ˤơLTkƤYϤǤd˺Ϥ碌٤ĉ䤷Ƥޤ
https://gdlc.connpass.com/event/128748/
Deep Learning_kФˤꡢо`बֱ椷n}ȡνQȤ_k֧Ԯץåȥե`KAMONOHASHIB餷ޤ
This document introduces HDInsight and big data concepts, detailing the Apache Hadoop ecosystem and its components such as HDFS and MapReduce. It discusses the role of HDInsight as a Hadoop-based service on Microsoft Azure, highlighting its scalability and integration with various tools. Additionally, it provides resources for further learning and insight into working with HDInsight clusters.
2019/5/16Ф줿GPU Deep Learning Community #11ˤơLTkƤYϤǤd˺Ϥ碌٤ĉ䤷Ƥޤ
https://gdlc.connpass.com/event/128748/
Deep Learning_kФˤꡢо`बֱ椷n}ȡνQȤ_k֧Ԯץåȥե`KAMONOHASHIB餷ޤ
This document introduces HDInsight and big data concepts, detailing the Apache Hadoop ecosystem and its components such as HDFS and MapReduce. It discusses the role of HDInsight as a Hadoop-based service on Microsoft Azure, highlighting its scalability and integration with various tools. Additionally, it provides resources for further learning and insight into working with HDInsight clusters.
2017/9/7 db tech showcase Tokyo 2017(JPOUG in 15 minutes)ˤưkݤǤ
SQLkФ˰餦IWӤϡߥåƥ륷ƥǤ꤬܆}ΤҤȤĤǤ
SQLޤȤưkФꡢIζضȤϤ뤳ȤǤиٻܤǤǤ???
APOӋ܆}Τᡢ_k̤νKPˤƤόIyȤࡩޤ
Τ褦״rˤơɤΤ褦ʸֶΤΤOracle˽hޤ
GPU ηؤΏäʤɤλAgMȥ饦ɤαkĤռ˰餤⤬ʹȤʹrܤʥޥ`ʹr늡ޥۡӥͥץꥱ`ʤɤȤΤ AI dƤȤ졢һΥǩ`ƥȤäƤ߶ȤʷѧΥե``⥨ɥ``ʹʤˤ٤ʤޤ
һǡAI ѧȤ~iޤ AI 뤢꤭ǥץȤʼޤꡢĿĤʧD줹褦ʥ`䡢yYˤĤƤԤˤĤhĤYޤäǤʤ褦ʥ`Ҋ褦ˤʤäƤޤ
ؤΥߥʩ`ǤϡAI ߶ȤʷˤĤƤ¥ȥɤȡʹɤˤĤơgHUYʤɤ̤ޤޤ
The document summarizes Spark developments from 2014-2015, including major distribution support from Cloudera, Hortonworks, and MapR. Community and committer growth increased substantially over the last year. Upcoming features for Spark 1.6 include a typed Dataset API, integration of DataFrames with GraphX and streaming, and performance enhancements through the Tungsten engine.
2. Copyright ?2017 CREATIONLINE, INC. All Rights Reserved
ԼB ľ r ߤĤȤ
ꥨ`饤ʽ Data Engineering Team - Director,
˥륿
ݺߣshare: http://www.slideshare.net/mkiuchi4
Nĸ
a. gihyo.jp: Mesosphere DCOSǤĤ륯饦ɥץꥱ`
b. սU饦ɥե`20166 Azure IoT Suiteu
c. Codezine: Cеѧ饦ɤXY BluemixApache Spark
Ǯʥ`ǩ`ϴ
Nv
a. Developer Summit 2016 Summer
b. սUBP ѥ֥å饦ɌBv
c. Cloudweek Hokkaido 2015/2016
d. оѧԺѧ ѧgΥ٩`о`
ѧgΥ٩`ߤΤοѧ`ץե`ࡹ
TҰApache Mesos, Apache Spark, ɢԥ`ƥ, 饦ɥ
`ƥ, NoSQL DB, DB
Oreilley Certified Developer on Apache Spark
2
18. Copyright ?2017 CREATIONLINE, INC. All Rights Reserved
Relational
KNOWS
ANDREAS
TOBIAS
MICA
DELIA
Person FriendPerson-Friend
ANDREAS
DELIA
TOBIAS
MICA
18
Graph
19. Copyright ?2017 CREATIONLINE, INC. All Rights Reserved 19
Connectedness and Size of Data Set
ResponseTime
Relational and
Other NoSQL
Databases
Neo4j
Neo4j is
1000x faster
Reduces minutes
to milliseconds
23. Copyright ?2017 CREATIONLINE, INC. All Rights Reserved
Neo4jC
23
Causal Clustering(3.1~, Enterprise Edition)
Paxos٩`Υޥޥ`饹
ץꥱ`ˌ
24. Copyright ?2017 CREATIONLINE, INC. All Rights Reserved
Neo4jC
24
Kerberos Authentication
New Cypher Query Execution
(up to 300% performance improvement)
Composite Index
28. Copyright ?2017 CREATIONLINE, INC. All Rights Reserved
Neo4j PoC ֧Ԯ`ӥ
28
? Neo4j֩`ȥ
C ȥ`˥ǥ⥢ץ
C ӥͥn}˄tƥ
C 1Lg
? Neo4jץꥱ`_k
?Neo4jǩ`֧Ԯ
C gs1|Node5|Relationship
C g3~4
28
29. Copyright ?2017 CREATIONLINE, INC. All Rights Reserved
Neo4j PoC ֧Ԯ`ӥ
29
? ?ϥȥ`˥
C CypherλAäޤ
C ץꥱ`APIǥ
C ե르ꥺࣨՓȌg
? åͼg֧Ԯ
C Neo4jävХåɥݩ`Ȥṩ
29