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
The Node.js Scalability Myth
Search
Sponsored
·
Ship Features Fearlessly
Turn features on and off without deploys. Used by thousands of Ruby developers.
→
Felix Geisendörfer
March 16, 2012
Programming
18k
16
Share
The Node.js Scalability Myth
Presentation given at the BerlinJS user group on March 15, 2012.
Felix Geisendörfer
March 16, 2012
More Decks by Felix Geisendörfer
See All by Felix Geisendörfer
tus.io - Resumable File Uploads (Lightning Talk)
felixge
2
820
Programming flying robots with JavaScript
felixge
2
1k
Programming flying robots with JavaScript
felixge
0
650
Programming an AR Drone Firmware with JS (de)
felixge
1
650
Faster than C?
felixge
1
1.3k
Flying robots over a 10.000 mile distance with JavaScript.
felixge
0
540
Faster than C?
felixge
1
700
The power of node.js (with quadcopters)
felixge
0
540
Faster than C?
felixge
0
470
Other Decks in Programming
See All in Programming
How Swift's Type System Guides AI Agents
koher
0
310
Road to RubyKaigi: Play Hard(ware)
makicamel
1
480
Going Multiplatform with Your Android App (Android Makers 2026)
zsmb
2
460
AIを導入する前にやるべきこと
negima
2
290
ハーネスエンジニアリングにどう向き合うか 〜ルールファイルを超えて開発プロセスを設計する〜 / How to approach harness engineering
rkaga
24
15k
「話せることがない」を乗り越える 〜日常業務から登壇テーマをつくる思考法〜
shoheimitani
4
880
How We Benchmarked Quarkus: Patterns and anti-patterns
hollycummins
1
160
의존성 주입과 모듈화
fornewid
0
150
NakouPAY説明用
annouim0
0
270
Don't Prompt Harder, Structure Better
kitasuke
0
790
TiDBのアーキテクチャから学ぶ分散システム入門 〜MySQL互換のNewSQLは何を解決するのか〜 / tidb-architecture-study
dznbk
1
190
2026_04_15_量子計算をパズルとして解く
hideakitakechi
0
120
Featured
See All Featured
How To Stay Up To Date on Web Technology
chriscoyier
790
250k
Dominate Local Search Results - an insider guide to GBP, reviews, and Local SEO
greggifford
PRO
0
160
Exploring anti-patterns in Rails
aemeredith
3
330
Faster Mobile Websites
deanohume
310
31k
[RailsConf 2023 Opening Keynote] The Magic of Rails
eileencodes
31
10k
The Curse of the Amulet
leimatthew05
1
12k
エンジニアに許された特別な時間の終わり
watany
106
240k
AI Search: Where Are We & What Can We Do About It?
aleyda
0
7.4k
How to Build an AI Search Optimization Roadmap - Criteria and Steps to Take #SEOIRL
aleyda
1
2k
Evolution of real-time – Irina Nazarova, EuRuKo, 2024
irinanazarova
9
1.3k
Neural Spatial Audio Processing for Sound Field Analysis and Control
skoyamalab
0
270
How to train your dragon (web standard)
notwaldorf
97
6.6k
Transcript
The Node.js Scalability Myth Felix Geisendörfer 15.03.2012 at BerlinJS user
group
(@)felixge(.de)
transloadit.com (using node since v0.0.6)
core contributor
I need to build an über-scalable web service. It will
be the next big thing!
Easy! I will use node.js and MongoDB!
Easy! I will use node.js and MongoDB!
Because node.js is “scalable”, right?
None
Audience?
Backend Developers?
Are you measuring app performance in production?
1000 req / s
100 req / s
10 req / s
10 requests / per second = 864000 requests / day
10 req / s
Scalability
None
The Node.js Scalability Myth • Threads don’t scale • Event
loops do
The Node.js Scalability Myth • Threads don’t scale • Event
loops do (Probably) not true in 2012 (Very likely) not relevant
More relevant • Understanding scalability • Using the right tool
for the job
Vertical Scalability
Adding more resources to a single node
None
Vertical Scalability • CPU • GPUs • Memory • Disk
• Network
CPU
Computers of the future will have hundreds of cores!
CPU • Node is single threaded (runs on a single
CPU) • Using multiple CPUs requires multiple node processes (usually via prefork) • No shared memory
CPU (v8) • V8 compiles JS to Assembly • Just-in-time
compilation (JIT) • Does reasonably well in those language “benchmarks”
GPUs
GPUs • No support in node itself • node-cuda addon
by Kashif Rasul
Memory
Memory • No hard memory limit on 64 bit (since
node-0.6 / v8-3.6.5) • JS is a garbage collected language (avoid huge heaps) • Buffers do not count towards heap
Network
Network • Node’s network stack is a thin layer on
top of underlaying system calls • Good at fully saturating network interfaces
Disk
Disk • Done in thread pool • Unfortunately along with
DNS at this point • Throughput ok, but not ideal yet
Horizontal Scalability
Adding more nodes to a system
Horizontal Scalability • Node has no horizontal scaling specific features,
but ... • Fast networking & JSON / Buffers -> DIY
Horizontal Scalability • Simple stack (requires only build tools, python)
• Deployable without container server • Good package management via npm
Horizontal Solution “I have the full solution ...”
Horizontal Solution “... but unfortunately it did not fit on
this slide”
tl;dr • Know your requirements • There are no silver
bullets
Questions? Slides will be available at felixge.de
Thanks