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
Thanosってどうですか?/Getting Started with Thanos
Search
watawuwu
July 06, 2018
Technology
1
1k
Thanosってどうですか?/Getting Started with Thanos
watawuwu
July 06, 2018
Tweet
Share
More Decks by watawuwu
See All by watawuwu
Prometheusでデータの水平分割を試みる/Let's split prometheus data
watawuwu
0
11k
データルーター?Vector/Getting Started with Vector
watawuwu
6
1.1k
KubernetesでWebアプリケーションをリリースするまでに必要なものは/What you need with Kubernetes
watawuwu
10
1.8k
Argo入門/Getting Started with Argo
watawuwu
0
1.1k
Concourse入門 / Concourse Getting Started
watawuwu
3
2.2k
Other Decks in Technology
See All in Technology
AWS CDK 実践的アプローチ N選 / aws-cdk-practical-approaches
gotok365
6
710
フィンテック養成勉強会#54
finengine
0
170
急成長を支える基盤作り〜地道な改善からコツコツと〜 #cre_meetup
stefafafan
0
120
エンジニア向け技術スタック情報
kauche
1
250
プロダクトエンジニアリング組織への歩み、その現在地 / Our journey to becoming a product engineering organization
hiro_torii
0
130
AWS Summit Japan 2025 Community Stage - App workflow automation by AWS Step Functions
matsuihidetoshi
1
250
Wasm元年
askua
0
140
Navigation3でViewModelにデータを渡す方法
mikanichinose
0
220
Fabric + Databricks 2025.6 の最新情報ピックアップ
ryomaru0825
1
130
A2Aのクライアントを自作する
rynsuke
1
170
Oracle Audit Vault and Database Firewall 20 概要
oracle4engineer
PRO
3
1.7k
Welcome to the LLM Club
koic
0
160
Featured
See All Featured
[RailsConf 2023] Rails as a piece of cake
palkan
55
5.6k
JavaScript: Past, Present, and Future - NDC Porto 2020
reverentgeek
48
5.4k
10 Git Anti Patterns You Should be Aware of
lemiorhan
PRO
657
60k
Rails Girls Zürich Keynote
gr2m
94
14k
GitHub's CSS Performance
jonrohan
1031
460k
The MySQL Ecosystem @ GitHub 2015
samlambert
251
13k
Keith and Marios Guide to Fast Websites
keithpitt
411
22k
Facilitating Awesome Meetings
lara
54
6.4k
GraphQLとの向き合い方2022年版
quramy
47
14k
Build your cross-platform service in a week with App Engine
jlugia
231
18k
How to train your dragon (web standard)
notwaldorf
92
6.1k
What’s in a name? Adding method to the madness
productmarketing
PRO
23
3.5k
Transcript
Getting Started with Thanos
profile: name: Wataru Matsui orz: [ zlab.co.jp, 3bi.tech ] twitter:
@watawuwu role: [ Server, Infra, SRE ]
Agenda ⊸ What’s Thanos ⊸ Why try Thanos & Key
features ⊸ Design ⊸ Demo ⊸ Conclusions
What’s Thanos?
HA metric system with unlimited storage capacity for
What’s Prometheus
Monitoring system & time series database (時間が少ないので省略)
Why try Thanos? And Key Features
メトリクスデータを オブジェクトストレージ に保存できる Why try Thanos? POINT!! Reliable metrics data
storage
複数Prometheusの メトリクスデータを マージし閲覧できる Why try Thanos? Global view POINT!!
ロングタームなクエリの応答性を維持 するため、メトリクスデータをダウンサ ンプリング Why try Thanos? Downsampling
依存コンポーネントが少なく 既存のPrometheus環境に 追加することが簡単 Why try Thanos? Seamless integration
Design
Querier(Thanos) Prometheus targets Sidecar(Thanos) Prometheus Sidecar(Thanos) targets targets targets Merge
& Deduplicate targets targets targets targets Global view HTTP API
SSD Prometheus Sidecar(Thanos) targets targets targets Blocks targets targets targets
Blocks Blocks targets targets targets Blocks targets targets targets Blocks Unlimited retention! Write
Unlimited retention! Read targets targets targets Blocks Cache Store(Thanos) targets
targets targets Blocks Querier(Thanos) Store API
targets targets targets Blocks Disk Compactor(Thanos) targets targets targets Blocks
Compaction & Downsampling Blocks Blocks
targets targets targets Blocks Querier(Thanos) Prometheus Sidecar(Thanos) Full Architecture Store(Thanos)
Compactor(Thanos)
Use Case
Only Backup Disk Prometheus Sidecar(Thanos) targets targets targets Blocks targets
targets targets Blocks Blocks targets targets targets Blocks targets targets targets Blocks
Querier(Thanos) Prometheus Sidecar(Thanos) Prometheus Sidecar(Thanos) HA or Aggregate
targets targets targets Blocks Querier(Thanos) Prometheus Sidecar(Thanos) All Store(Thanos) Compactor(Thanos)
Demo
None
Notice
Prometheus >= v2.0.0 可能であればv2.2.1
Conclusions
2018/07/04時点 Thanos has no official releases yet
まだまだ動作検証中! (カジュアルに使う分には問題ない)
導入前には本当に長期間の保存 が必要か確認!? • 不要なデータを削減 • スクラップ間隔と保存期間を見直す • KPIデータは別のストレージを検討 • 古いデータの価値(YAGNI)
場合によってはPrometheusだけでも十分!
Thanks! WE ARE HIRING!! Kubernetes, Cloud Native zlab.co.jp