Upgrade to Pro — share decks privately, control downloads, hide ads and more …

AWS 콘서트

Avatar for Daehyun Kim Daehyun Kim
October 11, 2012
Technology
0
500

AWS 콘서트

Amazon Web Services 소개. Amazon EC2, S3, EMR

Avatar for Daehyun Kim

Daehyun Kim

October 11, 2012
Tweet

More Decks by Daehyun Kim

See All by Daehyun Kim
사내 Git 저장소 개발 사례
Avatar for Daehyun Kim hatemogi
6
1.1k
김대현
Avatar for Daehyun Kim hatemogi
1
150
EF2011
Avatar for Daehyun Kim hatemogi
2
89

Other Decks in Technology

See All in Technology
KMP の Swift export
Avatar for Koki Hirokawa kokihirokawa
0
330
自動テストのコストと向き合ってみた
Avatar for SmartHR 品質保証部 qa
0
180
Function calling機能をPLaMo2に実装するには / PFN LLMセミナー
Avatar for Preferred Networks pfn
PRO
0
940
生成AIとM5Stack / M5 Japan Tour 2025 Autumn 東京
Avatar for you（@youtoy） you
PRO
0
220
後進育成のしくじり〜任せるスキルとリーダーシップの両立〜
Avatar for mtskhs matsu0228
7
2.5k
英語は話せません！それでも海外チームと信頼関係を作るため、対話を重ねた2ヶ月間のまなび
Avatar for ニイオカ niioka_97
0
120
Large Vision Language Modelを用いた 文書画像データ化作業自動化の検証、運用 / shibuya_AI
Avatar for Sansan R&D sansan_randd
0
110
関係性が駆動するアジャイル──GPTに人格を与えたら、対話を通してふりかえりを 習慣化できた話
Avatar for リリカル mhlyc
0
130
多野優介
Avatar for 多野優介 tanoyusuke
1
450
綺麗なデータマートをつくろう_データ整備を前向きに考える会 / Let's create clean data mart
Avatar for BrainPad brainpadpr
2
190
Modern_Data_Stack最新動向クイズ_買収_AI_激動の2025年_.pdf
Avatar for Sagara sagara
0
220
神回のメカニズムと再現方法/Mechanisms and Playbook for Kamikai scrumat2025
Avatar for moriyuya moriyuya
4
570

Featured

See All Featured
Large-scale JavaScript Application Architecture
Avatar for Addy Osmani addyosmani
514
110k
Context Engineering - Making Every Token Count
Avatar for Addy Osmani addyosmani
5
190
Improving Core Web Vitals using Speculation Rules API
Avatar for Sergey Chernyshev sergeychernyshev
19
1.2k
Building Better People: How to give real-time feedback that sticks.
Avatar for Will Jessup wjessup
368
20k
Making the Leap to Tech Lead
Avatar for Ryan Cromwell cromwellryan
135
9.5k
Mobile First: as difficult as doing things right
Avatar for Swwweet swwweet
224
10k
Side Projects
Avatar for Sacha Greif sachag
455
43k
Why You Should Never Use an ORM
Avatar for John Nunemaker jnunemaker
PRO
59
9.6k
BBQ
Avatar for Matthew Crist matthewcrist
89
9.8k
JavaScript: Past, Present, and Future - NDC Porto 2020
Avatar for David Neal reverentgeek
52
5.6k
The Art of Programming - Codeland 2020
Avatar for Erika Heidi erikaheidi
56
14k
Speed Design
Avatar for Sergey Chernyshev sergeychernyshev
32
1.1k

Transcript

  1. "NB[PO8FC4FSWJDFT ֙ਘ ӣ؀അ

  2. ߊ಴ղਊ ௿ۄ਋٘ஹೊ౴ "NB[PO8FC4FSWJDFTࣗѐ &$43%4&.3 ਋ܻоഝਊೡ݅ೠನੋ౟ ࢎղীبੑغݶજਸ݅ೠѪ

  3. ੉ߊ಴ܳٛҊ য૵ݶ ೧Ѿоמೠޙઁٜ ৌब൤

  4. ҾӘ೧ৈ  ҾӘೞݶਗ ౠ੿दр؀ী౟ې೗੉ށ۰ࢲࢲߡо؊೙ਃೠؘ ੌद੸ਵ۽పझ౟ೡѐߊࢲߡо೙ਃೠؘߊ઱߉ইঠೞա ౵ੌਸ੷੢ೞݶ ౟ې೗Ҋ޹হ੉ਢࢲ࠺झೡࣻহա ߔসೞҊࠁҙೞח؊ಞܻೠߑߨ਷হਸө ߎীࢲߡલযبਫ਼ખಞ൤ੜࣻחহա ইפ

    ؊աইоࢲ੢গоউզࣻחহա

  5. ௿ۄ਋٘ஹೊ౴

  6. ௿ۄ਋٘ஹೊ౴ &MBTUJD$BQBDJUZ 'BTUFSUJNFUPNBSLFU /P$BQ&Y 1BZBTZPVHP QBZGPSXIBUZPVVTF 'PDVTPOZPVSCVTJOFTT

  7. "NB[PO8FC4FSWJDFT "84חஹೊ౴ੋ೐ۄܳ௿ۄ਋٘ࢲ࠺झ )JHIMZSFMJBCMF 4DBMBCMF -PXDPTU ੹ࣁ҅աۄ੄ࣻभ݅࠺૑פझীੋ೐ۄ೒ۖಬઁҕ

  8. "84ઁಿ߂ࣛܖ࣌

  9. "84੹୓ઁಿ߂ࣛܖ࣌ IUUQBXTBNB[PODPNLPQSPEVDUT

  10. "84઱ਃઁಿ߂ࣛܖ࣌ "NB[PO&$ &MBTUJD-PBE#BMBODJOH "VUP4DBMJOH "NB[PO&.3 "NB[PO4 "NB[PO&#4 "NB[PO3%4 "NB[PO%ZOBNP%# "NB[PO424

    "NB[PO4&4 "NB[PO$MPVE'SPOU "NB[PO&MBTUJ$BDIF "84%JSFDU$POOFDU "84&MBTUJD#FBOTUBML "84$MPVE'PSNBUJPO "NB[PO$MPVE8BUDI

  11. য়਋ि_੉݆਷Ѣ׮ঌইঠ೤פө /P-JTUFODBSFGVMMZ೙ਃೠѪ݅ ॳݶؾפ׮ #VU ־о"84ী؀೧ফӝೡٸӒ ઺ޖ঺ী؀೧ফӝೞחѪੋ૑ഛੋ ೡ೙ਃח੓णפ׮

  12. "84੄੹ࣁ҅࠙ನ

  13. "NB[PO&$3FHJPO "WBJMBCJMJUZ;POFT "; ֙ୡӝળѐ";

  14. $MPVE'SPOU 3PVUF &EHF-PDBUJPOT DNS Service Content Delivery Dallas St.Louis Miami

    Jacksonville Los

  15. Ӓۢ਋ܻաۄח

  16. о੢оө਍ب௡ө૑੄֎౟ਕ௼ࣘبח IUUQXXXDMPVEQJOHJOGP XHFUIUUQEBVNDMPVETBNB[POBXTDPNSUH[ QJOHQJOHTPPVNF

  17. "845PLZPө૑੄֎౟ਕ௼ࣘب

  18. None

  19. Ӓېࢲ਋ܻաۄীࢲח 5PLZP૑৉ॳݶؾפ׮ ӒܻҊ &EHF੿بחࢤӡࣻب੓૑ঋਸөਃ

  20. "NB[PO&$

  21. "NB[PO&$ "NB[PO&MBTUJD$PNQVUF$MPVE ௿ۄ਋٘੄о࢚ࢲߡ ࢜ࢲߡੋझఢझ঳Ҋࠗ౴ೞחؘࣻᐠࣗਃ पઁࢎਊೠ݅ఀ݅ਃӘ૑ࠛ &MBTUJD#MPDL4UPSF &#4 "VUP4DBMJOH &MBTUJD-PBE#BMBODJOH

  22. "NB[PO&$൱ᠭ਍৔୓ઁ 3FE)BU&OUFSQSJTF-JOVY 6CVOUV 'FEPSB %FCJBO 0SBDMF&OUFSQSJTF-JOVY 8JOEPXT4FSWFS ᄑ

  23. "NB[PO&$ੋझఢझਬഋ 128 64 32 16 8 4 2 1 1

    2 4 8 16 32 64 EC2 Compute Units (HP) Memory (GB) Small 1.7 GB, 1 EC2 Compute Unit 1 virtual core $0.085/0.12 Micro 613 MB Up to 2 ECUs (for short bursts) $0.02/0.03 Large 7.5 GB 4 EC2 Compute Units 2 virtual cores $0.34/0.48 Extra Large 15 GB 8 EC2 Compute Units 4 virtual cores $0.68/0.96 Hi-Mem XL 17.1 GB 6.5 EC2 Compute Units 2 virtual cores $0.50/0.62 Hi-Mem 2XL 34.2 GB 13 EC2 Compute Units 4 virtual cores $1.00/1.24 Hi-Mem 4XL 68.4 GB 26 EC2 Compute Units 8 virtual cores $2.00/2.48 High-CPU Med 1.7 GB 5 EC2 Compute Units 2 virtual cores $0.17/0.29 High-CPU XL 7 GB 20 EC2 Compute Units 8 virtual cores $0.68/1.16 Cluster GPU 4XL 22 GB 33.5 EC2 Compute Units, 2 x NVIDIA Tesla “Fermi” M2050 GPUs $2.10/2.60 Cluster Compute 4XL 23 GB 33.5 EC2 Compute Units $1.60/1.98 ©2011 Amazon Web Services May not be reused or redistributed without written permission

  24. "NB[PO&$ੋझఢझਬഋ߂оѺ 5 PLZP 7 JSHJOJB .&.$16 MPDBMEJTL оѺ दрਘ ಴ળ

    5 (DV( ݅ਗ ಴ળ 5 (DV( ݅ਗ ಴ળ 5 (DV( ݅ਗ ಴ળ 5 (DV( ݅ਗ ؀ਊ۝ݫݽܻ 5 (DV( ݅ਗ ؀ਊ۝ݫݽܻ 5 (DV( ݅ਗ ؀ਊ۝ݫݽܻ 5 (DV( ݅ਗ Ҋࢿמ$16 5 (DV( ݅ਗ Ҋࢿמ$16 5 (DV( ݅ਗ ௿۞झఠ(16 7 (DV( /7*%*"5FTMBt'FSNJu.(16Y ݅ਗ ֫਷*0 7 (DVY(44% ݅ਗ DV_()[0QUFSPOژח9FPO೐۽ࣁࢲࢿמ `  दрੌࢎਊӝળ ݅ਗ޷݅੺࢏

  25. "NB[PO&$оѺ੿଼ ੋझఢझਃӘ 0OEFNBOEJOTUBODFT 3FTFSWFEJOTUBODFT ֙ড੿ ֙ড੿ 4QPUJOTUBODFT झషܻ૑ਃӘ ֎౟ਕ௼ਃӘ

  26. ௿ۄ਋٘➝࠺ਊ੺х

  27. &$ੋझఢझоѺ࢚ࣁ ೞ٘ਝয ؘ੉ఠࣃఠ࢚ݶ࠺ਊ ੹ӝܐ ֎౟ਕ௼੉ਊܐ ਍৔ࢲ࠺झੋ۱ .&.$16 MPDBMEJTL оѺ दрਘ

    ಴ળ 5 (DV( ݅ਗ ಴ળ 5 (DV( ݅ਗ

  28. &$֙֙ড੿ਃӘઁ 5PLZP POEFNBOEࢶѾઁӘহ੉ दр׼੉ਊਃӘ݅૑ࠛ ড੿ઁୡӝ࠺ਊਸࢶѾઁೞҊ दр׼ਃӘਸೡੋ߉਺ MJHIUNFEJVNIFBWZ (DV( ೠ׳ਃӘ POEFNBOE

    ݅ਗ ֙ড੿ ݅ਗ ֙ড੿ ݅ਗ

  29. 6ࢲߡ੄ਘ࠺ਊ ઁݧ؀۽ о੿ ঌ۰૓6ࢲߡ੄਍৔о੿࠺ਊ ࢚ݶױо ੹ӝױо ֎౟ਕ௼ನ౟࠺ਊਘ݅ਗ ೞ٘ਝযо੿࠺ਊ  

      ֙ѐਘਘ݅

  30. ਘ࠺ਊ࠺Ү AWS OnDemand AWS 1֙ড੿ AWS 3֙ড੿ 1U 16 20

    31 58

  31. ੺ডೞ۰ݶ ౠ੿दрীԙ೙ਃೠ݅ఀ੄ࢲߡܳࢎਊೞҊ߈ժ ӝࠄࢲߡ੗ਗਸ֙ড੿ ৘࢚ػೖ௼ఋ੐ਊ੗ਗਸ֙ড੿ਵ۽MJHIUড੿ઁࢲߡ ৘࢚஖ޅೠ౟ې೗ૐоחPOEFNBOEࢲߡܳӒٸӒٸ

  32. ؊੺ডೞ۰ݶ Leverage All Three Models 0 1000 2000 3000 4000

    5000 6000 7000 Reserved Instances On Demand Spot ©2011 Amazon Web Services May not be reused or redistributed without written permission

  33. ౟ې೗؀਽࠺ਊ

  34. য૵ݶ੿੘઺ਃೠ࠺ਊ਷ ੋ۱࠺ਊ %FW0QT ਍৔ੋ۱ ࠺ਊ ੄୭ࣗച ੢গ࠺ਊ

  35. ௿ۄ਋٘࠺ਊ ऴо ੿ഛೠёҙ੸࠺ਊ࠺Үח൨ٜ׮ா߄ா /P੢࠺࠺ਊ੉ԙऱ૑חঋणפ׮ :FTੌदрղղ೙ਃೠ੢࠺ੋоਃ

  36. "NB[PO&#4

  37. "NB[PO&#4 "NB[PO&MBTUJD#MPDL4UPSF #MPDLMFWFMTUPSBHFWPMVNF &$ੋझఢझীোѾ೧ࢎਊ (#_5# ಴ળࠅܬ೐۽࠺ઊ*014ࠅܬ ୭؀ *014 झշࢫࠂઁ 4۽ղࠁղӝоઉয়ӝ

  38. "NB[PO4

  39. "NB[PO4 "NB[PO4JNQMF4UPSBHF4FSWJDF ݒ0CKFDU ֙PSоਊࢿ ੋૐ ঐഐച )5514 #JU5PSSFOU೐۽ష௒ $MPVE'SPOU۽$%/ࢲ࠺झܳࣚऔҊчऱѱഝਊ

  40. ੹ࣁ҅4੷੢0CKFDUࣻ Amazon S3 Adoption Rate: Billions of Objects Stored 2.9

    14 40 102 262 566 0 100 200 300 400 500 600 Q4 2006 Q4 2007 Q4 2008 Q4 2009 Q4 2010 Q3 2011 Peak Requests: 370,000+ per second Doubled in 9 months! ֙2ӝળ রPCKFDUT੷੢ؽ

  41. 4ݒ۱ನੋ౟ ੉੿بन܉بݶ ޺Ҋ੷੢೧بغਗ਼ই ߹بߔস੉೙ਃೡө ➝ࢎۈ੄पࣻח )5514۽௿ۄ੉঱౟о૒੽੽Ӕ ౟ې೗ѣ੿዗

  42. "NB[PO4ࠗоӝמ 6TFSEFGJOFENFUBEBUB 4FSWFSTJEFFODSZQUJPO "DDFTTMPHHJOH 0CKFDUFYQJSBUJPO 0CKFDUWFSTJPOJOH 334 3FEVDF3FEVOEBODZ4UPSBHF

  43. "NB[PO3%4

  44. "NB[PO3%4 "NB[PO3FMBUJPOBM%BUBCBTF4FSWJDF .Z42-0SBDMF42-4FSWFS ੗زࣗ೐౟ਝযಁ஖ ߔস੗زച %#झշࢫ ੗زഐझ౟Ү୓ ࠂઁ

  45. ଵҊ1PTUHSFTPO"84 IUUQFOUFSQSJTFECDPNDMPVEEBUBCBTF

  46. "NB[PO&.3

  47. "NB[PO&.3 "NB[PO&MBTUJD.BQ3FEVDF )BEPPQ௿۞झఠܳ&$ਤীࢲ૊दҳز )BEPPQ௿۞झఠࢸ੿ प೯ ઑ੿ѣ੿዗ :PVsWFHPUUIFEBUB ؘ੉ఋ৬.3௏٘݅ળ࠺ೞࣁਃ

  48. "843FGFSFODF "SDIJUFDUVSFT

  49. None
  50. None
  51. None

  52. "84 ؀୓ޤоજ਷о 

  53. "84઱ਃݒ۱ನੋ౟ ਍৔֢۱୭ࣗച ੢গ؀࠺੗زച पઁ೙ਃ࠺ਊ݅૑ࠛ ೐۽Ӓې߁ઁযоמೠੋఠಕ੉झ ѐߊসޖী૘઺

  54. ੢গ؀࠺৬؀਽

  55. ࢎղഝਊоמࢿ

  56. ࢎղഝਊоמࢿ Ӗ۽ߥ೐۽ં౟೧৻ࢲ࠺झࣘبѐࢶ ੌद੸ѐߊप೷੢࠺ഝਊ ੿ӝ੸#BUDI1SPDFTTJOH ੸੿ӏݽ.BQ3FEVDF੘সীഝਊ Ҋࢿמ(16೐۽ࣁࢲഝਊ

  57. ࢎղഝਊоמࢿ ҅ࣘ ࢎ৻֎౟ਕ௼ীࢲ੄ݽפఠ݂ ౠ੿दр୶о౟ې೗؀਽ উ੹ೞҊ੷۴ೞѱߔসࠁҙ 4(MBDJFS /*4೐۽ં౟

  58. "84ԙॄঠ೤פө

  59. "84ഝਊ੄ࢶఖ ઺ࣗӝসীѱח୭Ҋ ੸੿ӏݽীѱחࢶఖ 4ח੸ӓॄࠅ݅ &.3਷ઙઙഝਊ &$חੌࠗ

  60. "84ࢎਊӝળо੉٘

  61. "84ࢎਊӝળо੉٘ ௿ۄ਋٘ӝࣿ౱ਤఃಕ੉૑ଵҊ

  62. ௿ۄ਋٘ӝࣿ౱

  63. ௿ۄ਋٘ӝࣿ౱ ࢎղ৻௿ۄ਋٘زೱ౵ঈ "84ܳನೣೠ௿ۄ਋٘ࢲ࠺झ੄ࢎղҕध଻օ ࢎղ"1*ా೤೒ۖಬઁҕ ࢎղ1BB4ઁҕ $BB4ѐߊᬊ &4BB4ѐߊ৘੿

  64. ݃ޖܻ "84ח௿ۄ਋٘҅੄ц "84о߈٘दч੉ऴѪ਷ইפ׮ ࢎղীഝਊ೧ࠅ݅ೠನੋ౟о੓׮

  65. хࢎ೤פ׮