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
高度に発達したシステムの異常は神の怒りと見分けがつかない / IPSJ-ONE 2017 y...
Search
Sponsored
·
SiteGround - Reliable hosting with speed, security, and support you can count on.
→
Yuuki Tsubouchi (yuuk1)
March 19, 2017
Technology
25k
3
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
高度に発達したシステムの異常は神の怒りと見分けがつかない / IPSJ-ONE 2017 y_uuki
IPSJ-ONE 2017 スライド。
Yuuki Tsubouchi (yuuk1)
March 19, 2017
More Decks by Yuuki Tsubouchi (yuuk1)
See All by Yuuki Tsubouchi (yuuk1)
「アラーティング」の話をしよう— SREconや論文等の最先端とのギャップをみる
yuukit
3
640
SAKURAONE:An Open Ethernet-based AI HPC System And Its Observed Workload Dynamicsin a Single-Tenant LLM Development Environment
yuukit
1
350
AIスーパーコンピュータにおけるLLM学習処理性能の計測と可観測性 / AI Supercomputer LLM Benchmarking and Observability
yuukit
1
940
SREはサイバネティクスの夢をみるか? / Do SREs Dream of Cybernetics?
yuukit
3
530
SREのためのテレメトリー技術の探究 / Telemetry for SRE
yuukit
13
3.6k
AIスパコン「さくらONE」の オブザーバビリティ / Observability for AI Supercomputer SAKURAONE
yuukit
2
1.5k
AIスパコン「さくらONE」のLLM学習ベンチマークによる性能評価 / SAKURAONE LLM Training Benchmarking
yuukit
2
1.1k
とあるSREの博士「過程」 / A Certain SRE’s Ph.D. Journey
yuukit
11
7k
eBPFを用いたAIネットワーク監視システム論文の実装 / eBPF Japan Meetup #4
yuukit
3
1.8k
Other Decks in Technology
See All in Technology
AIAU_UMEMOGU_ninomiya_slide
ninomiya_ii
0
230
AIのReact習熟度を測る
uhyo
2
650
脱SaaS!FDEを支えるプロビジョニングと分離設計
knih
0
240
iAEONの段階的リアーキテクト戦略 / iAEON's_Gradual_Re-architecture_Strategy
aeonpeople
0
230
Kiro Ambassador を目指す話
k_adachi_01
0
110
2026TECHFRESH畢業分享會 - AI 時代的人生存檔點
line_developers_tw
PRO
0
1.3k
徹底討論!ECS vs EKS!
daitak
0
150
Bucharest Tech Week 2026 - Guardians of the Cloud-Native Galaxy
edeandrea
PRO
0
120
失敗を資産に変えるClaude Code
shinyasaita
0
710
【NRUG vol.18】KubernetesにおけるNew Relicデータ取得量削減の考え方
nrug_member
0
160
ザ・データベース、MySQL ~ OSC 2026 Sendai ~
sakaik
0
130
AIネイティブな開発のサプライチェーンリスク対策 〜激動の開発現場でリスクに立ち向かう〜【ZennFes】
cscengineer
PRO
2
140
Featured
See All Featured
Amusing Abliteration
ianozsvald
1
210
Testing 201, or: Great Expectations
jmmastey
46
8.2k
Build your cross-platform service in a week with App Engine
jlugia
234
18k
How To Speak Unicorn (iThemes Webinar)
marktimemedia
1
490
The Art of Delivering Value - GDevCon NA Keynote
reverentgeek
16
2k
Mobile First: as difficult as doing things right
swwweet
225
10k
Digital Ethics as a Driver of Design Innovation
axbom
PRO
1
320
How Software Deployment tools have changed in the past 20 years
geshan
0
34k
It's Worth the Effort
3n
188
29k
A Guide to Academic Writing Using Generative AI - A Workshop
ks91
PRO
1
330
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
508
140k
JAMstack: Web Apps at Ludicrous Speed - All Things Open 2022
reverentgeek
1
480
Transcript
ߴʹൃୡͨ͠ γεςϜͷҟৗ ਆͷౖΓͱ ݟ͚͕͔ͭͳ͍ Πϯλʔωοτͱӡ༻ٕज़ݚڀձਪન גࣜձࣾͯͳ ௶༎थ
શࣗಈ γεςϜ
શࣗಈ γεςϜ
24࣌ؒ 365
ྗ
Infrastructure as Code Ϋϥυʹ͓͚Δαʔ όཧͷݪଇͱϓϥΫ ςΟε Photo by O’reilly Media
/ CC by 3.0 https://www.oreilly.co.jp/books/9784873117966/ https://creativecommons.org/licenses/by/3.0/deed
Kief Morris ஶ ٶԼ ߶ี ༁ ඌ ߴ߂ ༁ ʰInfrastructure
As Code Ϋϥυʹ͓͚ΔαʔόཧͷݪଇͱϓϥΫςΟεʱΑΓ Φʔτϝʔγϣϯπʔϧ͕Ͳ͏ ͍͏݁ՌΛੜΉ͔ʹ͍ͭͯࣗ৴ ͕࣋ͯͳ͍ͨΊɺΦʔτϝʔγϣ ϯπʔϧʹ͖ͤΓʹͳΔͷ ා͔ͬͨɻ 1.3.5 Φʔτϝʔγϣϯڪා
அ
ߴʹൃୡ
ࣗવք
ҟৗ
ະ
ਆͷౖΓ
؍ଌ
ࢹ Monitoring
ϔϞάϩϏϯ ݂ٿ ϔϚτΫϦοτ ന݂ٿ ݂খ൘ ݂ਗ਼ΧϧγϜ .$7 .$) .$)$ ૯ίϨεςϩʔϧ
)%-ίϨεςϩʔϧ -%-ίϨεςϩʔϧ தੑࢷ ۭෲ݂࣌ ऩॖظ݂ѹ ֦ுظ݂ѹ ͨΜͺ͘ જ݂ ૉૉ ΫϨΞνχϯ જ݂ ;55 "45 "-5 Ѝ(51 "-1 ૯ϏϦϧϏϯ ϩϏϦϊʔήϯ ૯ͨΜͺ͘ Ξϧϒϛϯ )#T߅ମ )#T߅ݪ )$7߅ମ Ξϛϥʔθ $31 ϦϚτΠυҼࢠ ݈߁அ
JOUFSGBDFFUIUY#ZUFT pMFTZTUFNYWEBVTFE MPBEBWH DQVVTFS DQVJEMF DQVTUFBM NFNPSZVTFE MJOVYTT4:/3&$7 MJOVY5*.&@8"*5 MJOVY6/$0//
MJOVYTT'*/8"*5 MJOVY$-04&8"*5 MJOVY-*45&/ JOPEFYWEBUPUBM DQVOJDF DQVTZTUFN DQVHVFTU NFNPSZCV⒎FST MJOVY$-04*/( MJOVY-"45"$, MJOVYTT'*/8"*5 NFNPSZGSFF JOUFSGBDFFUISY#ZUFT NFNPSZTXBQ@VTFE MJOVYDPOUFYU@TXJUDIFT MJOVYGPSLT MJOVY6/,08/ MJOVY-*45&/ MJOVYTT4:/4&/5 MJOVYTT&45"# NFNPSZDBDIFE DQVJPXBJU DQVTPGUJSR DQVIBSEJSR NFNPSZTXBQ@UPUBM NFNPSZUPUBM EJTLYWEBSFBET EJTLYWEBXSJUFT pMFTZTUFNYWEBTJ[F JOPEFYWEBGSFF ݈߁அ
σʔλ ϕʔε
ੑೳ VS ͓ۚ
)%% 44% ϝϞϦ
)%% 44% ϝϞϦ ͍ ;ͭ͏ ͍
)%% 44% ϝϞϦ ߴ͍ ;ͭ͏ ͍҆
ݹ͍σʔλ ΄ͱΜͲΞΫηε ͞Εͳ͍ͷͰ ͯ͘Α͍
)%% 44% ϝϞϦ ݹ͍ ৽͍͠ ͍ ͍҆
)%% 44% ϝϞϦ ݹ͍ ৽͍͠ ౷ ߹
DiamonDB https://github.com/yuuki/diamondb
ࢹ͚σʔλϕʔεੈͷதʹͨ͘ ͞Μ͋Δ Facebook͍ͭͬͯ͘Δ ΞΠσΞࣗମݹయత ͔͠͠ɺ ࢹ͚σʔλϕʔεʹద ༻͍ͯ͠Δྫͳ͍
ͯͳͷݱߦσʔλϕʔε ͱൺֱ͠ 100+ഒ ͷσʔλྔͷอ͕࣋ݱ࣮త ͳίετͰՄೳ
࣮ݧ
“PRINCIPLES OF CHAOS ENGINEERING”, http://principlesofchaos.org/ ΑΓ Chaos Engineering is the
discipline of experimenting on a distributed system in order to build confidence in the system’s capability to withstand turbulent conditions in production.
Θ͟ͱ ҟৗΛى͜͢
؍ଌ ࣮ݧ γεςϜ Ϟσϧ
؍ଌͱ࣮ݧʹΑΓ γεςϜಛੑΛ ֶश͠ͳ͕Β ࣗಈ࡞͢ΔγεςϜ
γεςϜཧऀΛ ະͷڪΕ͔Β ղ์͍ͨ͠