C10k (Problem 10,000 connections) in different languages ​​/ platforms

---

UPD. The second version of the benchmark here: eric.themoritzfamily.com/websocket-demo-results-v2.html

---

Warning : if you have complaints about the benchmark and / or code, the benchmark is laid out on GitHub , which allows you to correct bugs or report bugs to the author.

Read more about the problem of 10,000 connections: ru.wikipedia.org/wiki/Problema_10000_ connections

How to solve the problem of 10,000 connections through web socketstack Erlang, Go, Haskell (Snap), Java (Webbit), Node.js (websocket) and Python (ws4py)?
')


During the benchmark, a new client is launched every millisecond. Once a second, each client sends a message with the current time to the server, and the server sends this message back.

Implementation	Connection time (average)	Delay (average)	Posts	Compounds	Connection timeouts
Erlang	865ms	17ms	2849294	10,000	0
Haskell (Snap)	168ms	227ms	1187413	4996	108
Java (Webbit)	567ms	835ms	1028390	4637	157
Go	284ms	18503ms	2398180	9775	225
Node.js (websocket)	768ms	42580ms	1170847	5701	4299
Python (ws4py)	1561ms	34889ms	1052996	4792	5208

Implementations on Python and Node.js fell on about half of the connections.

Despite the fact that Go was expected to tear Erlang in performance, the delay was much higher and the server did not process the request on 225 connections.

Java Webbit was a dark horse, but ultimately showed better results than Go (157 raw connections).

The most amazing thing is that gevent (Python) and node.js stopped working after 5001 and 4792 connections respectively, despite the fact that both were implemented specifically to solve the C10k problem.

A more detailed description, code and raw data is available on GitHub: github.com/ericmoritz/wsdemo/blob/master/results.md

UPD. Comparisons include more and more people; tests for Tornado (Python) have appeared

---

UPD. The second version of the benchmark here: eric.themoritzfamily.com/websocket-demo-results-v2.html

---