Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
1日100個以上のHadoopクラスターを使い捨てる方法 / How to throw awa...
Search
Hiroyuki Nakazato / 中里 浩之
March 14, 2019
Technology
0
4.5k
1日100個以上のHadoopクラスターを使い捨てる方法 / How to throw away 100 Hadoop clusters a day
Hiroyuki Nakazato / 中里 浩之
March 14, 2019
Tweet
Share
More Decks by Hiroyuki Nakazato / 中里 浩之
See All by Hiroyuki Nakazato / 中里 浩之
Data Engineering Study 21 - Microsoft Fabric
nakazax
0
650
Azure Databricks Learning Series #3 - Network Design
nakazax
0
640
Azure Databricks Learning Series #2 - Databricks SQL
nakazax
0
470
Microsoft Build 2022 Recap Party!! Azure のデータ & 分析サービス 注目アップデート / microsoft-build-2022-recap-azure-data-and-analytics
nakazax
0
650
Microsoft Build 2022 - Azure のデータ & 分析サービス 最新アップデート / Microsoft Build 2022 Updates on Azure Data and Analytics Services
nakazax
1
850
Architecture patterns of Azure Cosmos DB & Azure Synapse Analytics
nakazax
1
390
CY2021 Updates for ADF & Synapse & Cosmos
nakazax
0
77
Ignite 2020 Update Azure Synapse Analytics
nakazax
0
1.1k
Ignite 2020 Update - Azure DB for MySQL and Postgres - Azure Arc - and more
nakazax
2
380
Other Decks in Technology
See All in Technology
AIを駆使したゲーム開発戦略: 新設AI組織の取り組み / sge-ai-strategy
cyberagentdevelopers
PRO
1
130
わたしとトラックポイント / TrackPoint tips
masahirokawahara
1
240
Nix入門パラダイム編
asa1984
2
200
独自ツール開発でスタジオ撮影をDX!「VLS(Virtual LED Studio)」 / dx-studio-vls
cyberagentdevelopers
PRO
1
170
来年もre:Invent2024 に行きたいあなたへ - “集中”と“つながり”で楽しむ -
ny7760
0
430
Java x Spring Boot Warm up
kazu_kichi_67
2
480
IaC運用を楽にするためにCDK Pipelinesを導入したけど、思い通りにいかなかった話
smt7174
1
100
CyberAgent 生成AI Deep Dive with Amazon Web Services / genai-aws
cyberagentdevelopers
PRO
1
470
ガバメントクラウド先行事業中間報告を読み解く
sugiim
1
480
Emacs x Nostr
hakkadaikon
1
150
顧客が本当に必要だったもの - パフォーマンス改善編 / Make what is needed
soudai
24
6.6k
MAMを軸とした動画ハンドリングにおけるAI活用前提の整備と次世代ビジョン / abema-ai-mam
cyberagentdevelopers
PRO
1
110
Featured
See All Featured
Embracing the Ebb and Flow
colly
84
4.4k
CSS Pre-Processors: Stylus, Less & Sass
bermonpainter
355
29k
No one is an island. Learnings from fostering a developers community.
thoeni
19
3k
Fashionably flexible responsive web design (full day workshop)
malarkey
404
65k
Building Better People: How to give real-time feedback that sticks.
wjessup
363
19k
Performance Is Good for Brains [We Love Speed 2024]
tammyeverts
3
370
Rebuilding a faster, lazier Slack
samanthasiow
79
8.6k
Product Roadmaps are Hard
iamctodd
PRO
48
10k
Statistics for Hackers
jakevdp
796
220k
Optimizing for Happiness
mojombo
376
69k
A Tale of Four Properties
chriscoyier
156
23k
Principles of Awesome APIs and How to Build Them.
keavy
126
17k
Transcript
݄ ιϑτόϯΫגࣜձࣾதཬߒ೭ ݸҎ্ͷ )BEPPQΫϥελʔΛ ͍ࣺͯΔํ๏ )BEPPQ4QBSL$POGFSFODF+BQBO
ࣗݾհ தཬߒ೭ φΧβτώϩϢΩ ॴଐ ιϑτόϯΫגࣜձࣾ גࣜձࣾ"HPPQ ϙδγϣϯ όοΫΤϯυνʔϜϦʔμʔ
"845FDIOJDBM.FJTUFS
͜ΜͳهࣄΛॻ͍͍ͯ·͢ IUUQTRJJUBDPNOBLB[BYJUFNTFEEGBB
Ґஔใऩू ղੳ ਓͷྲྀΕ
.VTUͷϩάΛؒόον։࢝·Ͱʹ%8)ʹೖ 8BOU%8)ͷೖૣ͚Εૣ͍΄Ͳϕλʔ ػೳ໘ ඇػೳ໘ ϩάͷϘϦϡʔϜԯ݅5#Φʔμʔ ࠓޙϩάྔ͕૿͑ଓ͚ΔͨΊεέʔϥϏϦςΟ͕ඞཁ ίετޮٴͼ҆ఆੑ͕ߴ͍͜ͱ &5-ॲཧͷཁ݅
ϩάऩू γεςϜ 4 3FETIJGU ,JOFTJT +FOLJOTPO&$ %FWFMPQFST %BUB"OBMZTUT
&-# &$ ऩूɾੵ &5- ղੳ &$ &$ ':ΞʔΩςΫνϟʔ
ϩάऩू γεςϜ 4 3FETIJGU ,JOFTJT +FOLJOTPO&$ %FWFMPQFST %BUB"OBMZTUT
&-# &$ ऩूɾੵ &5- ղੳ &$ &$ ':ΞʔΩςΫνϟʔ εέʔϥϏϦςΟ ͳͲʹ՝
࣌ؒͷϩάΛೖྗɺ1ZUIPONVMUJQSPDFTTJOHͰ $16Λͬͯ%8)ʹϩʔυՄೳͳܗࣜʹม ͋Δͷରͷϩά͕एׯҟͳΔͨΊ ૿͑ଓ͚Δϩάʹରͯ͠ &$ͷεέʔϧΞοϓͰରԠ ࣌ͷ&5-ॲཧͷ࣮
DYMBSHF ϦϦʔε W$16(J#.FN ֹ݄ສԁ DYMBSHF Լ०ϦϦʔε W$16(J#.FN ֹ݄ສԁ
εέʔϧΞοϓ ಄ଧͪϦεΫ ίϯϐϡʔςΟϯά࠷దԽΠϯελϯελΠϓ
࣌&5-ॲཧ ࣌&5-ॲཧ ͋Δ࣌ؒͷॲཧ͕Ҿ͘ͱޙଓ͕٧·Δ
&$ 4 ࣌&5-ॲཧ &5-ॲཧΠϝʔδ
ϩάऩू γεςϜ 4 3FETIJGU ,JOFTJT +FOLJOTPO&$ %FWFMPQFST %BUB"OBMZTUT
&-# &$ ऩूɾੵ &5- ղੳ &$ &$ ':ΞʔΩςΫνϟʔ ՝ͷղܾʹணख
৽ΞʔΩςΫνϟʔߏ εέʔϧΞτϞσϧ 4QBSLPO&.3Λ࠾༻ ॲཧͷಠཱੑ ࣌ؒͷ&5-ॲཧΛ Ϋϥελʔ͕୲ ͍ࣺͯՄೳͳ &.3όονॲཧج൫
ϩάऩू γεςϜ 4 3FETIJGU +FOLJOTPO&$ %FWFMPQFST %BUB"OBMZTUT &5-
ղੳ ऩूɾੵ ʜ ʜ 4QBSLPO&.3 )JWFPO&.3 ʜ 4 -BNCEB ,JOFTJT &-# &$ ৽ΞʔΩςΫνϟʔ (MVF
৽ΞʔΩςΫνϟʔͷϙΠϯτ &.3ͷεςοϓػೳͷ׆༻ -BNCEBʹΑΔ&.3Ϋϥελʔͷ੍ޚ ϑϧϚωʔδυ)JWFϝλετΞͷ׆༻
ΞυϗοΫੳ༻&.3Ϋϥελʔͷఏڙ
&.3ͷεςοϓػೳͷ׆༻ 1 2 3 4 εςοϓҙͷॲཧΛॱ൪ʹ࣮ߦ͢ΔΈ શεςοϓྃޙʹΫϥελʔΛऴྃ͢Δͷ੍ޚ͕ ؆୯ʹߦ͑Δ
݅ذ܁Γฦ͕͠ͳ͍γϯϓϧͳόονॲཧϑϩʔ Ͱ͋ΕεςοϓػೳͰेΧόʔՄೳ
&5-ॲཧͷεςοϓ &.3 ຊ൪ࢿࡐ +BS 4IFMM ςετࢿࡐ 4IFMM +BS
ੜϩά 1BSRVFU &5-ॲཧ༻ͷࢿࡐΛ4͔Βμϯϩʔυ TEJTUDQͰॲཧରͷϩάΛू TQBSLTVCNJUͰσʔλՃͱ1BSRVFUมΛ࣮ࢪ 1 2 3
-BNCEBʹΑΔΫϥελʔ੍ޚ 1 2 3 4 &.3Ϋϥελʔىಈ༻ͷ-BNCEBؔΛ࣮ 1ZUIPOͰ3VO+PC'MPXΛίʔϧ 3VO+PC'MPXύϥϝʔλʔ͕ඇৗʹଟ͍
ઃఆϑΝΠϧϥΠϒϥϦ)0$0/Ͱཧ͠࠶ར༻ੑΛ্ 1MBZ'SBNFXPSLͳͲͰ༗໊ͳ-JHIUCFOE͕ࣾ։ൃ ϩάऩूγεςϜͳͲͰ1MBZ'SBNFXPSLΛ׆༻ͯ͠ ͓Γ)0$0/ʹ׳Ε͍ͯͨͷ͕࠾༻ཧ༝
&5-༻Ϋϥελʔىಈϑϩʔ 1 2 3 4 Name = hourly-etl
ReleaseLabel = emr-5.20.0 Steps = [ { spark-submit hourly-etl.jar ${YYYYMMDD}/${HH} } ... IPVSMZDPOG IPVSMZDPOG )0$0/ύʔε 3VO+PC'MPX࣮ߦ
&5-ॲཧͷϦΧόϦʔ ޭ ޭ ࣦഊ ͕ҟৗऴྃ 1 2 3
4 ϦΧόϦ༻ ىಈ༻ Ϋϥελʔͷঢ়ଶΛ ఆظతʹνΣοΫ ىಈ༻-BNCEBΛ ݺͼग़͠ ࣦഊͨ࣌ؒ͠ͷ ΫϥελʔΛىಈ
ϑϧϚωʔδυ)JWFϝλετΞ 1 2 3 4 "84(MVF%BUB$BUBMPH <&.3 ରԠ>
Ҏલ.Z42-PO3%4Λ)JWFϝλετΞͱͯ͠ར༻ 3%4͕410'ʹͳΒͳ͍Α͏ʹؾΛݣ͏ඞཁ͕͋ͬͨ ΫϥελʔΛେྔʹىಈ͢ΔͱଓͰ͖ͳ͍FUD (MVF%BUB$BUBMPHϑϧϚωʔδυ͔ͭಉ࣌ଓ੍ݶͳ͠
ΞυϗοΫੳ༻Ϋϥελʔ ΤϯδχΞΞφϦετ͕ ؆୯ʹ&.3ΫϥελʔΛىಈՄೳʹ 1 2 3 4
Ϋϥελʔͷམͱ͠Εࢭ 1 2 3 4 I I I
ࢹ༻ Ϋϥελʔͷ Քಇ࣌ؒΛνΣοΫ 4MBDL௨ BEIPDDMVTUFS
ίετͷ ຖ࣌ͷ&5-ॲཧҎલΑΓ҆Ձʹ εϙοτΠϯελϯεΛੵۃతʹ׆༻͍ͯ͠ΔͨΊ 4શମͷίετ૿Ճ ΞυϗοΫੳ͕ؾܰʹͰ͖ΔͨΊੳ݁Ռͷσʔλ͕ 4ʹੵ͍͍ͯͬͯ͠Δ ෆཁͳσʔλΛఆظతʹআ͢ΔΑ͏ͳΈ͕ඞཁ
ϩάऩू γεςϜ 4 3FETIJGU +FOLJOTPO&$ %FWFMPQFST %BUB"OBMZTUT &5-
ղੳ ऩूɾੵ ʜ ʜ 4QBSLPO&.3 )JWFPO&.3 ʜ 4 -BNCEB ,JOFTJT &-# &$ ࠶ܝ৽ΞʔΩςΫνϟʔ (MVF
THANK YOU! ANY QUESTIONS?