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
SRE vs ラマダン
Search
Takayuki WATANABE (渡辺 喬之)
October 14, 2017
Technology
0
250
SRE vs ラマダン
Lightning talk at chukenweb
Takayuki WATANABE (渡辺 喬之)
October 14, 2017
Tweet
Share
More Decks by Takayuki WATANABE (渡辺 喬之)
See All by Takayuki WATANABE (渡辺 喬之)
[Money Forward x Shippio] BaySide-Tech-Nite (May 19, 2023)
takanabe
0
170
[Developers Summit 2023] ソフトウェアテスト新時代の幕開け: 機械学習とデータサイエンスで実現するテスト運用の高度化
takanabe
26
11k
SRE NEXT 2022: Sensible Incident Management for Software Startups
takanabe
2
7.8k
SRE NEXT 2020 [C6] Designing fault-tolerant microservices with SRE and circuit breaker centric architecture
takanabe
1
8.1k
Challenges for Global Service from a Perspective of SRE 2nd season
takanabe
3
4.9k
Practical Approaches to Achieve Continuous Deployment with Kubernetes
takanabe
0
260
Challenges for Global Service from a Perspective of SRE
takanabe
3
3.4k
Building infrastructure on AWS with Ruby
takanabe
0
380
Other Decks in Technology
See All in Technology
データ戦略部門 紹介資料
sansan33
PRO
1
3.8k
物体検出モデルでシイタケの収穫時期を自動判定してみた。 #devio2025
lamaglama39
0
270
OpenTelemetry が拡げる Gemini CLI の可観測性
phaya72
2
1.9k
Kubernetes self-healing of your workload
hwchiu
0
380
QA業務を変える(!?)AIを併用した不具合分析の実践
ma2ri
0
110
AWS UG Grantでグローバル20名に選出されてre:Inventに行く話と、マルチクラウドセキュリティの教科書を執筆した話 / The Story of Being Selected for the AWS UG Grant to Attending re:Invent, and Writing a Multi-Cloud Security Textbook
yuj1osm
1
130
MCP ✖️ Apps SDKを触ってみた
hisuzuya
0
300
混合雲環境整合異質工作流程工具運行關鍵業務 Job 的經驗分享
yaosiang
0
140
Dify on AWS 環境構築手順
yosse95ai
0
110
AI時代、“平均値”ではいられない
uhyo
8
2.3k
Databricks AI/BI Genie の「値ディクショナリー」をAmazonの奥地(S3)まで見に行く
kameitomohiro
1
380
旅で応援する✈️ NEWTが目指すコミュニティ支援とあたらしい旅行 / New Travel: Supporting by NEWT on Your Journey
mii3king
0
140
Featured
See All Featured
Art, The Web, and Tiny UX
lynnandtonic
303
21k
How to Ace a Technical Interview
jacobian
280
24k
Unsuck your backbone
ammeep
671
58k
Code Review Best Practice
trishagee
72
19k
Save Time (by Creating Custom Rails Generators)
garrettdimon
PRO
32
1.7k
Leading Effective Engineering Teams in the AI Era
addyosmani
7
580
Evolution of real-time – Irina Nazarova, EuRuKo, 2024
irinanazarova
9
990
Docker and Python
trallard
46
3.6k
Put a Button on it: Removing Barriers to Going Fast.
kastner
60
4k
Rails Girls Zürich Keynote
gr2m
95
14k
The Language of Interfaces
destraynor
162
25k
Designing Experiences People Love
moore
142
24k
Transcript
43&WTϥϚμϯ DIVLFOXFC
• ลڤ೭ [Takayuki Watanabe] • takanabe • @takanabe_w • ΫοΫύουגࣜձࣾ
• ΠϯϑϥετϥΫνϟʔ෦ SRE άϧʔϓ • ओʹGlobal αʔϏεͷΠϯϑϥͷ։ൃͱӡ༻Λ୲ ࣗݾհ
ΫοΫύουͷάϩʔόϧల։ʹͯ SRE ͕ܦݧͨ͠ ”ϥϚμϯ”ͱ͍͏Πϕϯτ͕ΫοΫύουͷαʔϏε Πϯϑϥʹ༩͑ΔӨڹʹ͍ͭͯ ࠓͷ͓
None
https://cookpad.com ࠃͷΫοΫύουͷτοϓϖʔδ
(MPCBMαʔϏεͷઆ໌
(MPCBMαʔϏεͷઆ໌ ΫοΫύουͷ Global αʔϏε (ҎԼ Global αʔϏε)
(MPCBMαʔϏεͷઆ໌ 21 ݴޠɾ67 ΧࠃҎ্Λରʹ αʔϏεΛల։ https://cookpad.com/us https://cookpad.com/id ɾ ɾ ɾ
(MPCBMαʔϏεͷઆ໌ ੈքதͷΠϕϯτͷӨڹΛड͚Δ
(MPCBMαʔϏεͷઆ໌ Global αʔϏεʹͱͬͯҰେΠϕϯτ ͷҰͭʹͳ͍ͬͯΔͷ͕ɺ
Ramadan
ϥϚμϯͬͯͳʹʁ • ώδϡϥྐྵͱΠεϥϜࣾձͰΘΕΔྺ๚ • ϥϚμϯώδϡϥྐྵʹ͓͚Δ9݄ͷ͜ͱ • அ৯ͷ͜ͱͰͳ͍ • ϥϚμϯظؒத ͷग़͔Β·ͰͷؒɺϜεϦϜ
ͷٛͷҰͭͱͯ͠அ৯͕ߦΘΕΔ • ΠεϥϜྐྵଠӄྐྵ • ϥϚμϯଠཅྐྵ͔ΒݟΔͱຖ11΄Ͳૣ·Δ
ϥϚμϯظؒதͷϢʔβϦΫΤετͷมԽ
ظؒதͷϦΫΤετͷมԽ ϥϚμϯظؒதͷϢʔβϦΫΤετͷมԽ !!
ΫοΫύουʹͱͬͯͷϥϚμϯ • ϥϚμϯظؒதɺޙՈͰීஈΑΓྉཧΛ͢Δਓ͕૿͑Δ • ΫοΫύουΛΒͳ͍ΠεϥϜݍͷਓʑ͕αʔϏεΛར༻ • ৽نϢʔβ֫ಘͷػձ • ΞΫςΟϒϢʔβ૿Ճ •
ϨγϐͷӾཡɾߘ͕૿Ճ • ϓϥοτϑΥʔϜͱͯ͠ͷັྗUP • ϓϨϛΞϜαʔϏεར༻ऀͷ૿Ճ • ऩӹͷ૿Ճ
43&ʹͱͬͯͷϥϚμϯ • զʑஅ৯Λ͠ͳ͍ • Ϣʔβ͔ΒͷϦΫΤετ͕͋ΔಥવഒʹͳΓͦΕ͕Ұϲ݄ଓ͘ • αʔόͷෛՙ૿Ճ • ීஈఆ͍ͯ͠ͳ͍ࣄͷൃੜ •
ෛՙ͕֬อ͍ͯ͠ΔαʔόϦιʔεΛ৯͍ਚ͘͢ • αʔόͷϩά͕ٸ૿ͯ͠σΟεΫΛຒΊਚ͘͢ • ීஈͳ͍εϩʔΫΤϦͰDB͕٧·Δ • etc …
ීஈൃੜ͠ͳ͍αʔόΞϥʔτ͕ൃੜ͕ͪ͠
43&WTϥϚμϯ • ϥϚμϯΛ • Ұ෦ͷٕज़ελοΫͷೖΕସ͑ • खಈͰରԠ͍ͯͨ͠ΦϖϨʔγϣϯΛࣗಈԽ • αʔό܈ͷΩϟύγςΟͷݟ͠ •
ܭըతʹϓϩμΫτ։ൃνʔϜͱ࿈ܞ • ύϑΥʔϚϯεͷվળґཔ • ظؒதʹ࣮ࢪ͢Δ͖ΘͲ͍ࢪࡦͷαϙʔτ • Ϣʔβʹීஈͱಉ͡ʹΑ͏ʹαʔϏεΛఏڙ͢ΔͨΊʹͨ͜͠ͱ • ͜ΕΒΛ࣮ࢪ͢Δ͜ͱͰظؒதͷ on call ରԠΛݮΒ͢
Ұ෦ͷٕज़ελοΫͷೖΕସ͑ • Database • MySQL on Amazon RDS (Magnetic) ͔ΒAmazon
Aurora ʹશ໘ೖΕସ͑ • CDN • Akamai͔Β Fastly ͷΓସ͑ • ΞϓϦέʔγϣϯͷ։ൃڥ • ৽نΞϓϦͷ։ൃ Docker + ECS Λར༻ ˠαʔϏεͷϨεϙϯελΠϜՄ༻ੑͷվળ
ΦϖϨʔγϣϯͷࣗಈԽ • αʔόϦιʔεͷෆΛճආ͢Δ • ΦʔτεέʔϧରͷαʔόΛ֦େ • ෛՙʹԠͯࣗ͡ಈతʹαʔόΛεέʔϧΞτ • haproxy
nginx ͳͲͷίϯϑΟάαʔόͷݱঢ়ʹ ߹Θͤͯࣗಈߋ৽(consul + consul-template) • ϩάͷٸ૿ʹΑΔσΟεΫͷṧഭͱͦͷରԠΛճආ • logrotate ΛͬͯσΟεΫʹϩάΛͣ͞ɺ࣌Ͱ S3 ʹ͢ ͙ʹόοΫΞοϓ͢ΔํࣜΛ࠾༻
αʔό܈ͷΩϟύγςΟͷݟ͠
ϓϩμΫτ։ൃνʔϜͱͷ࿈ܞ
ϓϩμΫτ։ൃνʔϜͱͷ࿈ܞ • ύϑΥʔϚϯεͷϘτϧωοΫΛੳͯ͠վળ • μογϡϘʔυͷڞ༗ • ίʔυϨϕϧͰϘτϧωοΫͷڞ༗ɺվળ
ظؒதʹ࣮ࢪ͢Δ͖ΘͲ͍ࢪࡦͷαϙʔτ • ྫ͑ɺങऩͨ͠αΠτͷσʔλΛΫοΫύουͷσʔλϕʔεʹϝ ϯςφϯεແ͠ͰϚΠάϨʔγϣϯ͢Δ࣌ • 2016͜ΕʹΑΓϥϚμϯظؒதʹ࣌ؒͷαʔϏεఀࢭ͕2 ճൃੜ • ࠓϥϚμϯظؒʹσʔλϚΠάϨʔγϣϯͷ༧ఆ͕͋ͬͨ •
σʔλҠߦํ๏ͷܭըίʔυϨϏϡʔʹࢀՃ • αʔόϦιʔεεϩʔΫΤϦϩάͳͲͷ֤छใΛϞχλϦϯ ά͢ΔμογϡϘʔυͷఏڙ 4FFBMTPIUUQTXXXZPVUVCFDPNXBUDI WQ&9;QV),Z+D
43&WTϥϚμϯͷ݁Ռ
43&WTϥϚμϯͷ݁Ռ େ͖ͳͳ͘ϥϚμϯظؒऴྃ!
ॴײ • ϥϚμϯΛ࢝Ίͱͯ͠αʔϏεͷ Global ల։ͳΒͰͷ՝ ໘ന͞ͱૺ۰͢Δ • େมͳ͜ͱଟ͍͕ɺֶͼଟ͍ • ָ͠Έͳ͕ΒΓӽ͍͑ͨ
• XXX ʹ͚ͯΩϟύγςΟͷݟ͕͠ඞཁΈ͍ͨͳՔಇ͕ ൃੜ͠ͳ͍Α͏ʹγεςϜΛߏஙɾվળ͍ͯ͘͜͠ͱ͕ٻ ΊΒΕ͍ͯΔ(ͨΓલ͕ͩ)
·ͱΊ • GlobalαʔϏεʹͱͬͯϥϚμϯେࣄͳΠϕϯτ • ීஈͱൺֱͯ͠ഒͷنͰϢʔβ͕૿͑Δ • SRE1ϲ݄ଓ͘ߴෛՙΛΓΔͨΊͷࢪࡦΛଧͭ • ෆ͍ͯ͠ΔϦιʔε͢ •
ࣗಈԽ • ϓϩμΫτ։ൃνʔϜͱͷ࿈ܞ • ࠓޙಛʹରࡦΛ͠ͳ͍ͰΓΕΔΑ͏ʹ͢Δͷ͕ॏཁ