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
About the problem of DBMS choice & what to do i...
Search
Minsk MongoDB User Group
June 07, 2012
Programming
3
120
About the problem of DBMS choice & what to do if you have gone the wrong way | Roman Bugaev
Roman Bugaev
Meetup #4
Minsk MongoDB User Group
June 07, 2012
Tweet
Share
More Decks by Minsk MongoDB User Group
See All by Minsk MongoDB User Group
MongoDB by Chef | Yauhen Artsiukhou
bymongo
0
140
MongoDB at IronMQ | Alexander Kolesen
bymongo
0
850
Scaling MongoDB | Sergey Gavruk
bymongo
2
190
Event sourcing + CQRS + MongoDB | Alex Shkor
bymongo
1
650
How it works. Indexes | Kirill Duborenko
bymongo
5
290
Aggregation Framework | Mikhail Burtylev
bymongo
1
120
MongoDB 2.2: Release update + Roadmap | Alvin Richards
bymongo
1
110
Meetup#6 Intro | Alex Litvinok
bymongo
1
59
Deploying MongoDB on Amazon WS | Michael Karpitsky
bymongo
2
120
Other Decks in Programming
See All in Programming
DSPy入門 Pythonで実現する自動プロンプト最適化 〜人手によるプロンプト調整からの卒業〜
seaturt1e
1
600
LangChain4jとは一味違うLangChain4j-CDI
kazumura
1
160
SourceGeneratorのマーカー属性問題について
htkym
0
170
ふつうの Rubyist、ちいさなデバイス、大きな一年
bash0c7
0
740
AWS×クラウドネイティブソフトウェア設計 / AWS x Cloud-Native Software Design
nrslib
13
2.6k
nuget-server - あなたが必要だったNuGetサーバー
kekyo
PRO
0
200
エラーログのマスキングの仕組みづくりに役立ったASTの話
kumoichi
0
120
クライアントワークでSREをするということ。あるいは事業会社におけるSREと同じこと・違うこと
nnaka2992
1
320
15年目のiOSアプリを1から作り直す技術
teakun
1
610
開発ステップを細分化する、破綻しないAI開発体制
kspace
0
110
米国のサイバーセキュリティタイムラインと見る Goの暗号パッケージの進化
tomtwinkle
2
510
CSC307 Lecture 13
javiergs
PRO
0
310
Featured
See All Featured
Designing for humans not robots
tammielis
254
26k
個人開発の失敗を避けるイケてる考え方 / tips for indie hackers
panda_program
122
21k
Large-scale JavaScript Application Architecture
addyosmani
515
110k
Design in an AI World
tapps
0
160
Ruling the World: When Life Gets Gamed
codingconduct
0
160
Stewardship and Sustainability of Urban and Community Forests
pwiseman
0
130
Bridging the Design Gap: How Collaborative Modelling removes blockers to flow between stakeholders and teams @FastFlow conf
baasie
0
470
Navigating the Design Leadership Dip - Product Design Week Design Leaders+ Conference 2024
apolaine
0
220
From Legacy to Launchpad: Building Startup-Ready Communities
dugsong
0
170
The Psychology of Web Performance [Beyond Tellerrand 2023]
tammyeverts
49
3.3k
GraphQLの誤解/rethinking-graphql
sonatard
75
11k
Making the Leap to Tech Lead
cromwellryan
135
9.8k
Transcript
Bugaev Roman About the problem of DBMS choice & what
to do if you have gone the wrong way
.NET web-‐‑developer
SQL or NoSQL, that is the question!
Ñ Learning management system Ñ SCORM Ñ 1 million active users CyberU
Ñ No transactions Ñ No relations Ñ No datatypes (some) SimpleDB
Key → ALribute → Value select count(*) from cyberu_users where
RecordType = 'user'
Prons & Cons
Ñ Everything is indexed Ñ Distributed Ñ Replicated Ñ Completely hosted on Amazon system
BUT Ñ Latency specially if you’re not on ec2 (even between nodes) Ñ Only strings, lexicographical string only Ñ Huge amount of code
Ñ If you love writing code-‐‑-‐‑ really, truly love to
write code-‐‑-‐‑ you'ʹll love it enough to write as liLle of it as possible. And SQL can help you.
SQL is awesome! -‐‑ Structured Query Language -‐‑ ACID Atomicity,
Consistency, Isolation, Durability -‐‑ Predictable -‐‑ Schema -‐‑ Based on rational algebra -‐‑ Standardized
No, really, it'ʹs awesome! -‐‑ Hardened -‐‑ Free and commercial
choices -‐‑ MySQL, PostgreSQL, Oracle, DB2, MS SQL... -‐‑ Commercial support -‐‑ Tooling -‐‑ Everyone knows it -‐‑ It'ʹs mature!
None
Sellerhit
So... -‐‑ Some preLy big and important problems -‐‑
And brightest engineers in the world -‐‑ Who loooove to build stuff
Yesterday'ʹs problem of biggest guys Is today'ʹs
problem of startup in garage
Ñ What schema can be fine for a lot of
product’s types? E-‐‑commerce
Books
Music
Jeans
Ñ Flexible (and sane) Ñ Searchable Ñ Easy to use Ñ Fast MongoDB
Ñ Atomic single document writes Ó If you need atomic writes across
multi-‐‑document – don’t use MongoDB And what about $$?
Ñ Orders, Inventory (id, count) and Payment Transaction you
can store in SQL Ñ UI – MongoDB, ecomerce -‐‑ SQL Most boring SQL Scheme ever
Ñ Double check you choice of DBMS and tools
Ñ Do it fast Ñ Check your logic Ñ Enterprise design Ñ Id<User> != Id<Product> Ñ Unit tests Ñ Delete code Ñ Optimize your tools Ñ Be brave! Data migrations
Use case – first, DB – last © Anonymous
Neo4J – Graph database
{ Thx! Q? Bugaev Roman, @rbugaev