5 ms·

600k concurrent HTTP connections with Clojure and http-kit (2013)

42 points by Mitt 13y ago

VeejayRampay 13y ago
10 seconds until someone mentions this benchmark is somehow unfair and nonsensical :) Still a very informative, promising and well formatted article, props.
GameOfTrolls 13y ago
This is not real-world. I dismiss this.
waffle_ss 13y ago
This has been posted twice before[1][2]. It should be noted that this benchmark was done over localhost, i.e. packets did not actually leave the computer, so who knows what kind of optimizations were done by the kernel / networking stack. When looking at examples where packets actually leave the computer, at least one person achieved 1M concurrent connections with Erlang in 2008, and WhatsApp were hitting 2.8M concurrent connections with Erlang in a production system in 2012[4]. Of course memory usage was a lot higher in these examples, but at least they were exercising the networking stack. [1]: https://news.ycombinator.com/item?id=5127251 https://news.ycombinator.com/item?id=5127251 [2]: https://news.ycombinator.com/item?id=5474331 https://news.ycombinator.com/item?id=5474331 [3]: http://www.metabrew.com/article/a-million-user-comet-application-with-mochiweb-part-1 http://www.metabrew.com/article/a-million-user-comet-applica... [4]: http://www.erlang-factory.com/upload/presentations/558/efsf2012-whatsapp-scaling.pdf http://www.erlang-factory.com/upload/presentations/558/efsf2...
- thomasjay 13y ago
  What would be a good way to benchmark these?
qwerta 13y ago
Now write database which would feed this with data...
ecesena 13y ago
Has anyone similar data on EC2 instances? I'd be particularly interested in small/medium ones. (localhost+no database are fine, it's just for the sake of comparison)
swah 13y ago
OT: I didn't know this nice http server with unified API for websocket and long polling. Maybe this is the reason SockJS and Socket.IO aren't getting the attention I thought they would? People are just using websocket or long polling, ignoring the other fallbacks??

42 points by Mitt 13y ago

10 seconds until someone mentions this benchmark is somehow unfair and nonsensical :) Still a very informative, promising and well formatted article, props.

GameOfTrolls 13y ago

This is not real-world. I dismiss this.

waffle_ss 13y ago

This has been posted twice before[1][2]. It should be noted that this benchmark was done over localhost, i.e. packets did not actually leave the computer, so who knows what kind of optimizations were done by the kernel / networking stack. When looking at examples where packets actually leave the computer, at least one person achieved 1M concurrent connections with Erlang in 2008, and WhatsApp were hitting 2.8M concurrent connections with Erlang in a production system in 2012[4]. Of course memory usage was a lot higher in these examples, but at least they were exercising the networking stack. [1]: https://news.ycombinator.com/item?id=5127251 https://news.ycombinator.com/item?id=5127251 [2]: https://news.ycombinator.com/item?id=5474331 https://news.ycombinator.com/item?id=5474331 [3]: http://www.metabrew.com/article/a-million-user-comet-application-with-mochiweb-part-1 http://www.metabrew.com/article/a-million-user-comet-applica... [4]: http://www.erlang-factory.com/upload/presentations/558/efsf2012-whatsapp-scaling.pdf http://www.erlang-factory.com/upload/presentations/558/efsf2...

thomasjay 13y ago
What would be a good way to benchmark these?

qwerta 13y ago

Now write database which would feed this with data...

ecesena 13y ago

Has anyone similar data on EC2 instances? I'd be particularly interested in small/medium ones. (localhost+no database are fine, it's just for the sake of comparison)

swah 13y ago

OT: I didn't know this nice http server with unified API for websocket and long polling. Maybe this is the reason SockJS and Socket.IO aren't getting the attention I thought they would? People are just using websocket or long polling, ignoring the other fallbacks??