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
機械学習によるマーケット健全化施策を支える技術
Search
Hirofumi Nakagawa/中河 宏文
May 23, 2018
Programming
0
240
機械学習によるマーケット健全化施策を支える技術
Hirofumi Nakagawa/中河 宏文
May 23, 2018
Tweet
Share
More Decks by Hirofumi Nakagawa/中河 宏文
See All by Hirofumi Nakagawa/中河 宏文
IoTデバイスでMLモデルを動かす技術
hnakagawa
0
180
Kanazawa_AI.pdf
hnakagawa
0
190
メルカリ写真検索における Amazon EKS の活用事例と プロダクトにおけるEdgeAI technologyの展望
hnakagawa
5
9k
メルカリの写真検索を支えるバックエンド CCSE 2019 version
hnakagawa
0
320
メルカリ写真検索における Amazon EKS の活用事例
hnakagawa
6
29k
メルカリの写真検索を支えるバックエンド
hnakagawa
1
1.2k
Mercari ML Platform
hnakagawa
1
17k
mlct.pdf
hnakagawa
2
2.1k
メルカリのマーケット健全化施策を支えるML基盤
hnakagawa
10
9.1k
Other Decks in Programming
See All in Programming
Goで実践するドメイン駆動開発 AIと歩み始めた新規プロダクト開発の現在地
imkaoru
4
590
CSC509 Lecture 01
javiergs
PRO
1
430
プロダクト開発をAI 1stに変革する〜SaaS is dead時代で生き残るために〜 / AI 1st Product Development
kobakei
0
490
CSC509 Lecture 02
javiergs
PRO
0
400
開発者への寄付をアプリ内課金として実装する時の気の使いどころ
ski
0
350
CSC509 Lecture 04
javiergs
PRO
0
290
2025年版 サーバーレス Web アプリケーションの作り方
hayatow
23
25k
私達はmodernize packageに夢を見るか feat. go/analysis, go/ast / Go Conference 2025
kaorumuta
2
490
なぜあの開発者はDevRelに伴走し続けるのか / Why Does That Developer Keep Running Alongside DevRel?
nrslib
3
370
Playwrightはどのようにクロスブラウザをサポートしているのか
yotahada3
7
2.3k
After go func(): Goroutines Through a Beginner’s Eye
97vaibhav
0
230
SpecKitでどこまでできる? コストはどれくらい?
leveragestech
0
510
Featured
See All Featured
Designing Experiences People Love
moore
142
24k
Rails Girls Zürich Keynote
gr2m
95
14k
Connecting the Dots Between Site Speed, User Experience & Your Business [WebExpo 2025]
tammyeverts
9
580
Designing for Performance
lara
610
69k
Building Better People: How to give real-time feedback that sticks.
wjessup
368
20k
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
46
7.6k
Building an army of robots
kneath
306
46k
Designing for humans not robots
tammielis
254
25k
Art, The Web, and Tiny UX
lynnandtonic
303
21k
Improving Core Web Vitals using Speculation Rules API
sergeychernyshev
19
1.2k
Why You Should Never Use an ORM
jnunemaker
PRO
59
9.6k
Responsive Adventures: Dirty Tricks From The Dark Corners of Front-End
smashingmag
252
21k
Transcript
ػցֶशʹΑΔϚʔέοτ ݈શԽࢪࡦΛࢧ͑Δٕज़ Mercari Server Side Tech Talk Vol.2 ʙCREφΠτʙ
hnakagawa
ࣗݾհ • Hirofumi Nakagawa (hnakagawa) • 20177݄ೖࣾ • ॴଐSRE •
σόΠευϥΠό։ൃ͔Βϑϩϯ τΤϯυ։ൃ·ͰΔԿͰ • NOT MLΤϯδχΞ • https://github.com/hnakagawa
͓ࣄ • ML Platform։ൃ • MLΤϯδχΞͱSREͷεΩϧΪϟοϓΛຒΊ Δ • ML Reliability,
SysML?, MLOps? • SREͷཱ͔ΒCREνʔϜΛॿ͚Δ
ML Platform • ͷML Platform • kubernetesϕʔε • ϩʔΧϧڥͱΫϥελڥͷ ࠩΛநԽ͢Δ
• ศརAPI܈ • طଘͷML FrameworkΛ༻͠ ؆୯ʹTraining/ServingΛߦ͏ ڥΛఏڙ
ͦͷ͏ͪOSSͰެ։༧ఆ(ଟ
ࠓͷAgenda ϦΞϧλΠϜࢹγεςϜ
ϦΞϧλΠϜࢹγεςϜ • ௨শ Lovemachine • ML Platform্ʹ࣮͞Ε͍ͯΔ .-1MBUGPSN USBJOJOHDMVTUFS -PWFNBDIJOF
($4 GKE PubSub .-1MBUGPSN TFSWJOHDMVTUFS -PWFNBDIJOF
ML ModelͷServing….?
Model Serving APIͷߏྫ 5FOTPS'MPX 4FSWJOH 5' .PEFM 5' .PEFM 'MBTL
4, .PEFM 4, .PEFM 4, .PEFM gRPC .FSDBSJ"1* REST FlaskͰલॲཧΛߦ͍ ཪͷTensorFlow Servingʹ͍͛ͯΔ
Model Serving API Streaming ver ͷߏྫ 5FOTPS'MPX 4FSWJOH 5' .PEFM
5' .PEFM .-1MBUGPSN 'SBNFXPSL PS "QBDIF#FBN 4, .PEFM 4, .PEFM 4, .PEFM gRPC PubSub
TensorFlow Serving • TensorFlow project͕ఏڙͯ͠ ͍ΔServingڥ • PythonॲཧܥΛհͣ͞ʹTFͷ modelΛservingͰ͖Δ •
ඪ४ͷ࣮ͰgRPCͰAPIΛ ఏڙ
ModelͱίϯςφɾΠϝʔδ • ڊେͳML ModelΛίϯςφɾΠϝʔδʹؚΊ Δ͔൱͔ • ؚΊͳ͍ͷͰ͋ΕԿॲʹஔ͢Δ͔ • ϙʔλϏϦςΟੑͱϩʔυ࣌ؒͷτϨʔυΦϑ •
ྑ͍ΞΠσΟΞ͕͋Εڭ͑ͯԼ͍͞…
௨ৗͷAPIͱҧ͏ • ѻ͏ϦιʔεɺModelαΠζ͕େ͖͘ͳΔ ߹͕ଟ͍(ඦMBʙGB) • CPUɾϝϞϦϦιʔεͷফඅ͕ܹ͍͠ • ߹ʹΑͬͯGPU͏
ϝϞϦফඅ • LovemachineͷPython࣮෦࣮ߦ࣌ʹ 2GBϝϞϦΛফඅ͢Δˠࠓޙ͞Βʹ૿͑Δ༧ ఆ͋Δ • Scikit-learnͰهड़͞ΕͨTF-IDFͷલॲཧ෦ ͕େ͖͘ͳΔࣄ͕ଟ͍
Pythonͱฒྻੑ • વThread͕͑ͳ͍(GILͷͨΊ) • ϓϩηεຖʹModelΛϩʔυ͢Δͱඞཁͳϝ ϞϦαΠζ͕େ͖͘ͳΔˠ Blue-Green DeployͷোʹͳΔ
ਖ਼PythonͰͷServing Πϯϑϥతʹਏ͍ࣄ͕ଟ͍…
ϝϞϦΛݡ͘͏ • fork͢ΔલʹmodelΛϩʔυ͠Copy on Write Λޮ͔͢ • k8sͷone process per
containerηΦϦ͋ ͑ͯഁ͍ͬͯΔ
Copy On Writeͷ෮श ϝϞϦ ϓϩηε ࢠϓϩηε 2.fork 1BHF" 1.allocation ಉ͡ྖҬΛࢀর
ϓϩηε͕ϝϞϦͷ༰Λ ॻ͖͑Δͱ… ϝϞϦ ϓϩηε ࢠϓϩηε 1BHF" 1BHF# OS͕ผͷྖҬΛAllocationͯ͠ݩσʔλΛίϐʔ͢Δ ผͷྖҬΛࢀর
Current Issues • Mercari APIͱͷͭͳ͗ࠐΈʹۤ࿑ ˠ Ұ௨Γ࡞Εޙ࠶ར༻Ͱ͖Δͣ • ਓؒͷߦಈΛ૬खʹ͍ͯ͠Δҝɺσʔλͷ͕ม ΘΓ͔ͬͨ͢Γɺ༧֎ͷ͕ൃੜͨ͠Γͯ͠ɺ
ରԠ͠ଓ͚Δඞཁ͕͋Δ ˠ ML Model࡞ऀʹෛ୲ֻ͕͔Γଓ͚Δ ˠ SREͱͯࣗ͠ಈԽΛؚΜͩΈͰղܾ͍ͨ͠
Future Plans • ࣾͷσʔλ͔ΒಛྔΛநग़͢Δͯ͠ Embedding͢Δ൚༻ͷΈ ˠదͳྨثͱΈ߹ΘͤΕɺ୭Ͱͦͦ͜ ͜ͷྨϞσϧΛ࡞Ͱ͖Δ? →FBLearner Flowతͳͭ? •
ࣾͷղܾʹಛԽͨ͠ઐ༻ͷAutoMLతͳԿ ͔?
·ͱΊ • ML ModelͷServingʹɺগ͠௨ৗͱҧ͏Πϯϑ ϥ͕ඞཁʹͳΔ →·ͩϕετɾϓϥΫςΟε͔Βͳ͍ • ਓͷߦಈΛ૬खʹ͢Δͷେม • ͦͦMLͳػೳΛຊ֨ӡ༻͠Α͏ͱ͢Δͱɺେ
෯ͳࣗಈԽɾΈԽΛਐΊͳ͍ͱ্ख͘ߦ͔ͳ ͍
͝ਗ਼ௌ͋Γ͕ͱ͏͍͟͝·ͨ͠!!
We are Hiring!!
SRE ML Reliability • SysML? MLOps? ৽͍͠Job description • SREεΩϧ+MLͷجૅࣝ
• MLΠϯϑϥͷࣗಈԽɾΈԽΛਪ͠ਐΊͯ ͘ΕΔਓࡐ • ͪΖΜଞͷ৬छઈࢍืूத!!
ৄࡉͪ͜Β https://careers.mercari.com/