6 ms·
So all your application does is accept connections, send hello world, and close them again?
by lmb 10y ago
So all your application does is accept connections, send hello world, and close them again?
- OhSoHumble 10y agoAbsolutely! I provide a "hello world as a service" platform.
- Cthulhu_ 10y agoHow does it compare to https://github.com/salvatorecordiano/hello-world-as-a-service https://github.com/salvatorecordiano/hello-world-as-a-servic..., which is written in JS / Node? Does yours do more requests / second? Do you have an enterprise plan?
- OhSoHumble 10y agoI question that project's long-term viability. Event loop based languages inherently limit systems to inefficient and unreliable concurrency models. My platform, which I've received seed funding for, is entirely done in Erlang. This technology decision will better enable me to deliver more hello worlds per nanosecond than Node ever could.
- w8rbt 10y agoHWaaS
- amelius 10y agoThe theory is that also other programs will be affected, with a similar slowdown.
- andy_ppp 10y agoIt's a regression in the closing of the connection in the http server in the standard library. If really worried about performance, the fasthttp package is much faster than the one included in go. "In short, fasthttp server is up to 10 times faster than net/http." https://github.com/valyala/fasthttp https://github.com/valyala/fasthttp
- elithrar 10y ago(And doesn't support HTTP/2, so pick your poison)
- andy_ppp 10y agoI thought the whole point of HTTP/2 was to transfer as many things as possible in one connection so maybe this bug is less of a concern for these types of connection.
- buro9 10y agoThe great thing is that now fasthttp can be a bit slower and still make that claim.
- DougBTX 10y agoSure, but what does "similar slowdown" mean? The headline says "20%", which would imply that a "similar slowdown" for a 100 milliseconds response would bring it to 120 milliseconds. But what is actually happening is that every request takes roughly an extra 0.04 milliseconds, so the time might go up to 100.04 milliseconds, probably within margin of error for most services anyway.
- lsaferite 10y agoSeems like an easy debunk here is a test that has a more realistic response delay instead so that the slowdown can de demonstrated as an absolute amount vs. a percentage amount.
- lossolo 10y agoIt's so hard to imagine high traffic application with api serving information to clients from array that is refreshed once in 10 minutes? So basically most of the time you touch only this array on every request which is like sending hello world example. In this situation it will be noticeable if you have a lot of traffic but then you would not use stdlib http library but something like fasthttp which doesn't change that there are other real world use cases that would be affected, not a lot of them but they exist.