$30 off During Our Annual Pro Sale. View Details »

Build Image Classification service with Amazon ECS and GPU instances

Yuichiro Someya
November 22, 2016
Programming
4
2.3k

Build Image Classification service with Amazon ECS and GPU instances

Yuichiro Someya

November 22, 2016
Tweet

More Decks by Yuichiro Someya

See All by Yuichiro Someya
にんげんがさき 基盤はあと / Developers over ML platform
ayemos
0
10k
機械学習をスモールスタートさせる方法 / small machine learning
ayemos
3
1.7k
アットホームな分析基盤の作り方 / Homemade Machine Learning Toolkits
ayemos
1
770
サービス開発、機械学習、クラウド / the trinity of machine learning
ayemos
0
2.8k
成長を止めない機械学習のやり方 / Don't stop 'til you get enough (data).
ayemos
15
4.7k
AWS で加速する機械学習 / Accelerate Machine Learning with AWS
ayemos
1
290
クックパッドの機械学習基盤 2018 / Machine Learning Platform at Cookpad ~ 2018 ~
ayemos
15
16k
PyTorchとCaffe2とONNXと深層学習モデルのデプロイについて
ayemos
1
2.7k
クックパッドにおけるAWS GPUインスタンスの利用事例 / Powering by AWS GPU Instances in Cookpad Inc
ayemos
0
340

Other Decks in Programming

See All in Programming
Kaigi_on_Rails_2022スポンサーLT登壇資料.pdf
yuta_onishi_97
1
300
카프카, 대규모 클러스터 운영 후기
kakao
PRO
0
120
Elixirコミュニティ の歩き方〜国内オンライン編〜
elijo
3
930
ROS2自律走行実現に向けて2 / Autonomous driving realization by ROS2 test framework
fixstars
2
170
『良いコード／悪いコードで学ぶ設計入門』を一歩深める読み方 / deepen good code bad code
minodriven
12
4.1k
AWS Glueではじめるデータレイク
marchin1989
0
180
Google Cloud Spannerにおけるページネーションとインデックスの考え方
khmer495
0
120
A hitchhiker's guide to Kotlin MultiPlatform
aldefy
0
120
FrankenPHP: a modern app server for PHP and Symfony apps
dunglas
0
230
How to dynamically validate data with Symfony
marionleherisson
0
120
Composing an API with Kotlin vol 2 (Advanced Kotlin Dev Day 2022)
zsmb
1
260
なんでもPull Requestにする / Kichijoji.pm 31
utgwkk
1
1.6k

Featured

See All Featured
Become a Pro
speakerdeck
PRO
3
3.1k
Done Done
chrislema
176
14k
What's in a price? How to price your products and services
michaelherold
231
9.6k
RailsConf & Balkan Ruby 2019: The Past, Present, and Future of Rails at GitHub
eileencodes
120
29k
Writing Fast Ruby
sferik
613
57k
Dealing with People You Can't Stand - Big Design 2015
cassininazir
351
21k
Product Roadmaps are Hard
iamctodd
35
7.5k
VelocityConf: Rendering Performance Case Studies
addyosmani
317
22k
Side Projects
sachag
451
37k
Gamification - CAS2011
davidbonilla
75
4k
Fantastic passwords and where to find them - at NoRuKo
philnash
31
1.8k
Design by the Numbers
sachag
271
17k

Transcript

  1. Build Image Classification service with AWS ECS and GPU instances

    Yuichiro Someya @ Cookpad

  2. • છ୩ ༔Ұ࿠ [Yuichiro Someya] • ౦޻େେֶӃ ܭࢉ޻ֶઐ߈ म࢜ •

    '16 ৽ଔ @ ΫοΫύου • github.com/ayemos • twitter.com/kumasan_com echo `whoami`

  3. • ྉཧࣸਅͷࣗಈऩूαʔϏεΛ • CaffeNet[1]Λݩʹ࡞ͬͨྉཧը૾ೝࣝϞσϧͱ • Amazon SQS/S3 Ͱߏங͞Εͨσʔλϑϩʔͱ • Amazon

    ECS (GPU instance) Λར༻ͯ͠ӡ༻͍ͯ͠Δ࿩ <>IUUQTHJUIVCDPN#7-$DB⒎F Agenda

  4. • ྉཧࣸਅͷࣗಈऩूαʔϏεΛ • CaffeNet[1]Λݩʹ࡞ͬͨྉཧը૾ೝࣝϞσϧͱ • Amazon SQS/S3 Ͱߏங͞Εͨσʔλϑϩʔͱ • Amazon

    ECS (GPU instance) Λར༻ͯ͠ӡ༻͍ͯ͠Δ࿩ <>IUUQTHJUIVCDPN#7-$DB⒎F Agenda

  5. ΫοΫύου • Ϩγϐ਺ɿ 250ສ඼Ҏ্ • ݄࣍ؒར༻ऀ਺ɿ 6,000ສਓҎ্

  6. • εϚϗ಺ͷࣸਅ͔Βྉཧ͚ͩΛࣗಈతʹऩू • Ұ෦ͷϢʔβʔ޲͚ʹݶఆతʹެ։த ྉཧ͖Ζ͘

  7. • CaffeNetΛ ྉཧʗඇྉཧ ൑ఆ޲͚ʹFine Tuningͨ͠Ϟσϧ • Caffe[1]Ͱֶश͞ΕͨϞσϧΛChainerͷCaffe emulatorͰಡΉ
 ref: http://docs.chainer.org/en/stable/reference/caffe.html

    • ෼ྨΧςΰϦΛ ྉཧʗඇྉཧ ʹมߋ͠ɺΫοΫύου্ͷ
 ྉཧࣸਅΛ࢖ֶͬͯश <>IUUQDB⒎FCFSLFMFZWJTJPOPSH CookpadNet

  8. • CookpadNet͸Ͳ͜Ͱ൑ఆΛߦ͍ɺͦͷ݁Ռ͸Ͳ͜ʹͲ͏఻͑Δ ͷ͔ʁ • ൑ఆϞσϧΛΫϥΠΞϯτʹஔ͍ͯ൑ఆ • ϞσϧαΠζ͕େ͖͍(100MB~)ͷͰɺݱ࣮తͰͳ͍ • (αΠζͷখ͍͞ϞσϧΛݚڀத) •

    ൑ఆΛߦ͏ίϯϙʔωϯτΛ֎෦ʹஔ͘ • HTTP Serverʁ σʔλϑϩʔʗϫʔΫϑϩʔ

  9. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO4FSWFS QZUIPO DIBJOFS

  10. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ 1045DMBTTJGZ\QIPUPCJOBSZ^ $MBTTJpDBUJPO4FSWFS QZUIPO DIBJOFS

  11. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ 1045DMBTTJGZ\QIPUPCJOBSZ^ $MBTTJpDBUJPO4FSWFS QZUIPO 1045DMBTTJGZ\QIPUPCJOBSZ^ DIBJOFS

  12. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ 1045DMBTTJGZ\QIPUPCJOBSZ^ $MBTTJpDBUJPO4FSWFS QZUIPO 1045DMBTTJGZ\QIPUPCJOBSZ^ SFTVMU\JT@GPPECPPM^

    DIBJOFS

  13. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ 1045DMBTTJGZ\QIPUPCJOBSZ^ $MBTTJpDBUJPO4FSWFS QZUIPO 1045DMBTTJGZ\QIPUPCJOBSZ^ SFTVMU\JT@GPPECPPM^

    DIBJOFS SFTVMU\JT@GPPECPPM^

  14. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ 1045DMBTTJGZ\QIPUPCJOBSZ^ $MBTTJpDBUJPO4FSWFS QZUIPO 1045DMBTTJGZ\QIPUPCJOBSZ^ SFTVMU\JT@GPPECPPM^

    DIBJOFS ը૾ͷΞοϓϩʔυ ը૾ॲཧ ൑ఆ SFTVMU\JT@GPPECPPM^

  15. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ 1045DMBTTJGZ\QIPUPCJOBSZ^ $MBTTJpDBUJPO4FSWFS QZUIPO 1045DMBTTJGZ\QIPUPCJOBSZ^ SFTVMU\JT@GPPECPPM^

    DIBJOFS ը૾ͷΞοϓϩʔυ ը૾ॲཧ ൑ఆ SFTVMU\JT@GPPECPPM^ >>> 300~500 ms <<<

  16. • ը૾ॲཧͱϞσϧʹinferenceʹֻ͕͔ͦͦ࣌ؒ͜͜Δ
 (300~500ms) • APIαʔόʔ͔Βಉظతʹୟ͚ͳ͍
 (Unicorn ͷ worker͕ਚ͖ͯ͠·͏) • Amazon

    S3, SQSΛར༻ͨ͠ඇಉظͳ൑ఆॲཧϫʔΫϑϩʔ σʔλϑϩʔʗϫʔΫϑϩʔ

  17. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS "NB[PO4 4UPSBHF

    "NB[PO424 2VFVF %#

  18. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS "NB[PO4 4UPSBHF

    <6QMPBEQIPUPUPDMBTTJGZ> "NB[PO424 2VFVF %#

  19. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS "NB[PO4 4UPSBHF

    <6QMPBEQIPUPUPDMBTTJGZ> "NB[PO424 2VFVF FORVFVF
 \LFZ@PO@TTUSJOH^ %#

  20. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS "NB[PO4 4UPSBHF

    <6QMPBEQIPUPUPDMBTTJGZ> "NB[PO424 2VFVF FORVFVF
 \LFZ@PO@TTUSJOH^ EFRVFVF
 \LFZ@PO@TTUSJOH^ %#

  21. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS "NB[PO4 4UPSBHF

    <6QMPBEQIPUPUPDMBTTJGZ> "NB[PO424 2VFVF FORVFVF
 \LFZ@PO@TTUSJOH^ EFRVFVF
 \LFZ@PO@TTUSJOH^ <%PXOMPBE*NBHF> %#

  22. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS "NB[PO4 4UPSBHF

    <6QMPBEQIPUPUPDMBTTJGZ> "NB[PO424 2VFVF FORVFVF
 \LFZ@PO@TTUSJOH^ EFRVFVF
 \LFZ@PO@TTUSJOH^ 1045SFTVMU
 \LFZ@PO@TTUSJOH SFTVMU\JT@GPPE <%PXOMPBE*NBHF> %#

  23. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS "NB[PO4 4UPSBHF

    <6QMPBEQIPUPUPDMBTTJGZ> 1045JT@QIPUP\LFZ@PO@TTUSJOH^ "NB[PO424 2VFVF FORVFVF
 \LFZ@PO@TTUSJOH^ EFRVFVF
 \LFZ@PO@TTUSJOH^ 1045SFTVMU
 \LFZ@PO@TTUSJOH SFTVMU\JT@GPPE <%PXOMPBE*NBHF> %#

  24. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS SFTVMU\JT@GPPECPPM^ "NB[PO4

    4UPSBHF <6QMPBEQIPUPUPDMBTTJGZ> 1045JT@QIPUP\LFZ@PO@TTUSJOH^ "NB[PO424 2VFVF FORVFVF
 \LFZ@PO@TTUSJOH^ EFRVFVF
 \LFZ@PO@TTUSJOH^ 1045SFTVMU
 \LFZ@PO@TTUSJOH SFTVMU\JT@GPPE <%PXOMPBE*NBHF> %#

  25. $MJFOU "OESPJE J04 "1*4FSWFS SVCZ $MBTTJpDBUJPO8PSLFS QZUIPO DIBJOFS SFTVMU\JT@GPPECPPM^ "NB[PO4

    4UPSBHF <6QMPBEQIPUPUPDMBTTJGZ> 1045JT@QIPUP\LFZ@PO@TTUSJOH^ "NB[PO424 2VFVF FORVFVF
 \LFZ@PO@TTUSJOH^ EFRVFVF
 \LFZ@PO@TTUSJOH^ 1045SFTVMU
 \LFZ@PO@TTUSJOH SFTVMU\JT@GPPECPPM^^ <%PXOMPBE*NBHF> ඇಉظʹ൑ఆॲཧ

  26. • ྉཧࣸਅͷࣗಈऩूαʔϏεΛ • CaffeNet[1]Λݩʹ࡞ͬͨྉཧը૾ೝࣝϞσϧͱ • Amazon SQS/S3 Ͱߏங͞Εͨσʔλϑϩʔͱ • Amazon

    ECS Λར༻ͯ͠ӡ༻͍ͯ͠Δ࿩ <>IUUQTHJUIVCDPN#7-$DB⒎F Agenda

  27. • ECS: Amazon EC2 Container Service • Docker ContainerΛEC2Ͱߏ੒͞ΕͨΫϥελʹ഑ஔ(Task) •

    github.com/eagletmt/hako • ECSͷߏ੒ΛyamlϑΝΠϧͰ؅ཧ ECSͱGPUͱDockerͱ…

  28. "8471$ # cookpadnet-worker.yml scheduler: type: ecs region: ap-northeast-1 cluster: hako-production-g2

    desired_count: 1 app: image: cookpadnet-worker-gpu cpu: 128 memory: 3072 memory_reservation: 2048 env: AWS_REGION: ap-northeast-1 COOKPADNET_ENV: production ... %PDLFS3FHJTUSZ ։ൃऀ EPDLFSQVTI IBLPEFQMPZ &$4 EPDLFSQVMM 5BTL DPPLQBEOFUXPSLFS

  29. "8471$ # cookpadnet-worker.yml scheduler: type: ecs region: ap-northeast-1 cluster: hako-production-g2

    desired_count: 1 app: image: cookpadnet-worker-gpu cpu: 128 memory: 3072 memory_reservation: 2048 env: AWS_REGION: ap-northeast-1 COOKPADNET_ENV: production ... %PDLFS3FHJTUSZ ։ൃऀ EPDLFSQVTI IBLPEFQMPZ &$4 EPDLFSQVMM 5BTL DPPLQBEOFUXPSLFS DockerԽ͞ΕͨWorkerΛ
 hakoͰσϓϩΠ & ߏ੒؅ཧ

  30. w XPSLFSͰ͸(16Λ࢖༻ w ಉՁ֨ଳͷ$16Πϯελϯεͱൺ΂ͯ ഒͷੑೳࠩ w %PDLFS (16 GPU

  31. • Driver͕ඞཁ • nvidia-driverͷkernel module • ಉ͡όʔδϣϯͷuser-level drivers • Docker

    Container͔ΒGPU devicesΛૢ࡞͢Δҝ
 Containerʹద੾ͳLinux Capabilityͷઃఆ͕ඞཁ Ծ૝Խ v.s. Χʔωϧ

  32. ubuntu EPDLFSDPOUBJOFS ཧ૝ OWJEJB(16 VTFSMFWFMESJWFS LFSOFMNPEVMFT

  33. ubuntu EPDLFSDPOUBJOFS ཧ૝ OWJEJB(16 VTFSMFWFMESJWFS LFSOFMNPEVMFT

  34. ubuntu EPDLFSDPOUBJOFS ཧ૝ OWJEJB(16 VTFSMFWFMESJWFS LFSOFMNPEVMFT ESJWFSךQBUIכ04ח״׶殯ז׷

  35. NVIDIA Docker • Docker CLIͷബ͍ϥούʔ • `docker run` ࣌ʹඞཁͳvolumeΛࣗಈతʹmount
 ͯ͘͠ΕΔ

  36. NVIDIA Docker • Docker CLIͷബ͍ϥούʔ • `docker run` ࣌ʹඞཁͳvolumeΛࣗಈతʹmount
 ͯ͘͠ΕΔ

    "NB[PO&$4דכ劢؟ه٦ز

  37. ubuntu EPDLFSDPOUBJOFS ཧ૝ OWJEJB(16 VTFSMFWFMESJWFS LFSOFMNPEVMFT

  38. ubuntu EPDLFSDPOUBJOFS ཧ૝ OWJEJB(16 VTFSMFWFMESJWFS LFSOFMNPEVMFT (ಉҰόʔδϣϯ)

  39. ubuntu EPDLFSDPOUBJOFS ཧ૝ OWJEJB(16 VTFSMFWFMESJWFS LFSOFMNPEVMFT 㣐⡤鍑寸 (ಉҰόʔδϣϯ)

  40. • Driver͕ඞཁ • nvidia-driverͷkernel module • ಉ͡όʔδϣϯͷuser-level drivers • Docker

    Container͔ΒGPU devicesΛૢ࡞͢Δҝ
 Containerʹద੾ͳLinux Capabilityͷઃఆ͕ඞཁ Ծ૝Խ v.s. Χʔωϧ

  41. • GPUσόΠε͸ಛघͳϑΝΠϧͱͯ͠ଘࡏ • ΞΫηε͢ΔͨΊʹಛఆͷCapabilityઃఆ͕ඞཁ Ծ૝Խ v.s. Χʔωϧ EPDLFSSVOa EFWJDFEFWOWJEJBEFWOWJEJBa EFWJDFEFWOWJEJBVWNEFWOWJEJBVWNa

    HQVXPSLFS

  42. • GPUσόΠε͸ಛघͳϑΝΠϧͱͯ͠ଘࡏ • ΞΫηε͢ΔͨΊʹಛఆͷCapabilityઃఆ͕ඞཁ Ծ૝Խ v.s. Χʔωϧ EPDLFSSVOa EFWJDFEFWOWJEJBEFWOWJEJBa EFWJDFEFWOWJEJBVWNEFWOWJEJBVWNa

    HQVXPSLFS &$4ͷ5BTLఆٛʹ͓͍ͯEFWJDFΦϓγϣϯ͸ະαϙʔτ

  43. • GPUσόΠε͸ಛघͳϑΝΠϧͱͯ͠ଘࡏ • ΞΫηε͢ΔͨΊʹಛఆͷCapabilityઃఆ͕ඞཁ Ծ૝Խ v.s. Χʔωϧ EPDLFSSVOa EFWJDFEFWOWJEJBEFWOWJEJBa EFWJDFEFWOWJEJBVWNEFWOWJEJBVWNa

    HQVXPSLFS &$4ͷ5BTLఆٛʹ͓͍ͯEFWJDFΦϓγϣϯ͸ະαϙʔτ

  44. • GPUσόΠε͸ಛघͳϑΝΠϧͱͯ͠ଘࡏ • ΞΫηε͢ΔͨΊʹಛఆͷCapabilityઃఆ͕ඞཁ Ծ૝Խ v.s. Χʔωϧ EPDLFSSVOQSJWJMFHFEHQVXPSLFS

  45. Ծ૝Խ v.s. Χʔωϧ EPDLFSSVOQSJWJMFHFEHQVXPSLFS • capability શ։์ • rootͰ࣮ߦ͞Ε͍ͯΔdockerd্ͷcontainerͷதͰrootΛ औ͍ͬͯΔͷͰ৭ʑग़དྷΔ

    EPDLFSSVOQSJWJMFHFEBMQJOFMBUFTUEBUFT • GPUσόΠε͸ಛघͳϑΝΠϧͱͯ͠ଘࡏ • ΞΫηε͢ΔͨΊʹಛఆͷCapabilityઃఆ͕ඞཁ

  46. • GPUσόΠε͸ಛघͳϑΝΠϧͱͯ͠ଘࡏ • ΞΫηε͢ΔͨΊʹಛఆͷCapabilityઃఆ͕ඞཁ Ծ૝Խ v.s. Χʔωϧ EPDLFSSVOQSJWJMFHFEHQVXPSLFS • rootҎ֎ͷϢʔβʔͰ࣮ߦ͢Δ͜ͱʹ͢Δ

    • DockerFile಺Ͱ `USER runner`

  47. • ྉཧࣸਅͷࣗಈऩूαʔϏεΛ • CaffeNet[1]Λݩʹ࡞ͬͨྉཧը૾ೝࣝϞσϧͱ • Amazon SQS/S3 Ͱߏங͞Εͨσʔλϑϩʔͱ • Amazon

    ECS (GPU instance) Λར༻ͯ͠ӡ༻͍ͯ͠Δ࿩ <>IUUQTHJUIVCDPN#7-$DB⒎F Agenda