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
How we learned to love the Data Center Operatin...
Search
DevOps Pro
June 01, 2016
Technology
0
230
How we learned to love the Data Center Operating System
by Saulius Valatka
DevOps Pro Vilnius 2016
DevOps Pro
June 01, 2016
Tweet
Share
More Decks by DevOps Pro
See All by DevOps Pro
Practical Application Logging and Monitoring
devopspro
0
110
Google Cloud Platform for DevOps Professionals
devopspro
0
300
Dev vs Ops transformation to Agility DevOps culture
devopspro
1
240
Using Test Automation for Continuous Integration, Delivery and Quality
devopspro
0
100
How the hell do I run my microservices in production, and will it scale?
devopspro
0
130
Accelerating innovation through AWS Cloud technology
devopspro
0
93
The Field Guide to Understanding Declarative Systems
devopspro
0
86
DevOps is the answer! What was the question again?
devopspro
0
160
From Monolith to Docker Distributed Applications
devopspro
0
260
Other Decks in Technology
See All in Technology
15 years with Rails and DDD (AI Edition)
andrzejkrzywda
0
200
AzureでのIaC - Bicep? Terraform? それ早く言ってよ会議
torumakabe
1
560
Greatest Disaster Hits in Web Performance
guaca
0
260
配列に見る bash と zsh の違い
kazzpapa3
3
160
名刺メーカーDevグループ 紹介資料
sansan33
PRO
0
1k
Bedrock PolicyでAmazon Bedrock Guardrails利用を強制してみた
yuu551
0
240
小さく始めるBCP ― 多プロダクト環境で始める最初の一歩
kekke_n
1
430
登壇駆動学習のすすめ — CfPのネタの見つけ方と書くときに意識していること
bicstone
3
110
データの整合性を保ちたいだけなんだ
shoheimitani
8
3.1k
Embedded SREの終わりを設計する 「なんとなく」から計画的な自立支援へ
sansantech
PRO
3
2.5k
モダンUIでフルサーバーレスなAIエージェントをAmplifyとCDKでサクッとデプロイしよう
minorun365
4
210
【Ubie】AIを活用した広告アセット「爆速」生成事例 | AI_Ops_Community_Vol.2
yoshiki_0316
1
100
Featured
See All Featured
Imperfection Machines: The Place of Print at Facebook
scottboms
269
14k
Gemini Prompt Engineering: Practical Techniques for Tangible AI Outcomes
mfonobong
2
280
Measuring Dark Social's Impact On Conversion and Attribution
stephenakadiri
1
130
The AI Revolution Will Not Be Monopolized: How open-source beats economies of scale, even for LLMs
inesmontani
PRO
3
3k
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
marcduiker
35
2.4k
Claude Code どこまでも/ Claude Code Everywhere
nwiizo
61
52k
Beyond borders and beyond the search box: How to win the global "messy middle" with AI-driven SEO
davidcarrasco
1
53
The Cult of Friendly URLs
andyhume
79
6.8k
SEOcharity - Dark patterns in SEO and UX: How to avoid them and build a more ethical web
sarafernandez
0
120
How STYLIGHT went responsive
nonsquared
100
6k
B2B Lead Gen: Tactics, Traps & Triumph
marketingsoph
0
55
Design in an AI World
tapps
0
140
Transcript
HOW WE LEARNED TO LOVE THE DATA CENTER OPERATING SYSTEM
SAULIUS VALATKA / ADFORM
Online Advertising Full Stack Platform
Online Advertising Full Stack Platform Realtime “smart“ ads Forecasting, fraud
detection, etc.
Online Advertising Full Stack Platform Realtime “smart“ ads Forecasting, fraud
detection, etc. 1mln QPS under 100ms 1TB daily data
EVOLUTION
MIDDLE AGES ctrtrain.ec2-aws.com test2.ec2-aws.com modelling.ec2-aws.com
THE TORTURE # yum install python R libboost-3.12 $ scp
script.R test.aws.com:/script.R # crontab -e
THE TORTURE # yum install python R libboost-3.12 $ scp
script.R test.aws.com:/script.R # crontab -e “strange, worked on my machine …”
RENAISSANCE ctrtrain.ec2-aws.com test2.ec2-aws.com worker-1.adform.com worker-2.adform.com ab34na3n ar2afga3n
None
CONTAINERIZE ! self contained artifacts isolated runtime basically no overhead
unified deployment
BUT WAIT … what about configuration ?
The twelve-factor app stores config in environment variables Env vars
are easy to change between deploys without changing any code There is little chance of them being checked into the code repo accidentally They are a language- and OS-agnostic standard
BUT WAIT … where do I log ? and what
about metrics ?
None
MODERN ERA e34sadf ab34na3n af4f5a4r aafde33a fa45daws faes4fa3 aaf444a2 fas3rfa4
None
MARATHON the init of the DCOS constraints deployment { "id":
“my-nginx", "container": { "type": "DOCKER", "docker": { "image": "nginx:1.7.7", "network": "BRIDGE", } }, "instances": 1, "cpus": 0.5, "mem": 128 }
SPRINT the exec of he DCOS will be open sourced
scheduler to follow!
MANAGING RESOURCES how much memory do I really need ?
MANAGING RESOURCES how much memory do I really need ?
and CPUs ? what does 0.5 CPUs mean anyway ?
MANAGING RESOURCES how much memory do I really need ?
and CPUs ? what does 0.5 CPUs mean anyway ? and what happens with the network ?
None
ISOLATION cgroups: cpu cpuset memory blkio net_cls
NETWORK ISOLATION Layer 3 routing software defined networks
ATOMIC AGE a4faw3f 4afsdgg asdf4faf se4faw aw3d3ff g4aefgsd 5gsdgr54s a4rff4afa
4f4qaf4
SERVICE DISCOVERY where is my app ? how do I
reach it ? won’t containers conflict about ports ?
MARATHON-LB
PERSISTENCE so .. where do I store my data ?
on the host ? won’t it disappear ?
PERSISTENCE / /opt/app/cache /var/lib/docker/devicemapper /var/lib/mesos/slave/volumes
PERSISTENCE / /opt/app/cache /opt/app/profile /var/lib/docker/devicemapper /var/lib/mesos/slave/volumes /mnt/sdc network block storage
FUTURE PLANS DC/OS IP per container Containerize all the things
@adforminsider