[blasdel]

> but where's the evidence what people see for active/total ratios in the real world?

Yes Zed, where the fuck is it? You're claiming SCIENCE! based on your worst-case synthetic localhost benchmarks, and then turning around and wildly guessing as to real-world performance characteristics with internet latencies.

Worse, your whole thesis hinges off of ATR but you made no effort to measure it anywhere, instead you're passive-aggressively berating us to do it.

[zedshaw]

Wow here we are again, you not reading my article. I ran the same test that everyone else runs for poll vs. epoll, then used R to craft graphs and tested hypothesis. It was not a localhost test.

So far all you've got is trolling HN comments. YOU WIN!

[jacquesm]

Pipes, localhost, who is counting, as far as I'm concerned that's the same thing, making it seem as if for the purpose of this test that's a significant difference is simply conversational trickery.

If you have tested this on real live servers then there is no evidence of that in your posting, and to suggest that this:

http://dpaste.de/32o8/

is anything but a localhost test is simply bogus.

The only use case where you may be right that poll is advantageous as far as I can see is streaming media servers (video, audio, other large files), image servers are the ones with the worst active-to-total ratios, especially if the images are small. I should know, I only serve up a few billion of them every day. A few years ago or so I was stupid enough to think that video was hard, man was I ever wrong. Repeated connections to the same host, that's a much bigger killer than pumping bits.

[zedshaw]

Testing this on real live servers is confounding. Man you guys really don't get this. If you want to test epoll and poll over file descriptors you test that. You don't test a billion other things in a network server. That confounds your results.

But what's really amazing is this is the test the proponents epoll have been using for 8 years. Where was your objection back when they were using it for that?

[blasdel]

That you didn't say anything about the machines you ran it on was a huge red flag. It looks like jacquesm has you nailed on the localhost testing, I hadn't noticed the link to the code before.

I'll trust that you accurately measured the ATR boundary between poll and epoll in your specific synthetic benchmark. That's then completely undermined by your handwaving in this thread about what ATR looks like in the wild, and the lack of any way for us to relate your microbenchmark with the real world.

[zedshaw]

The original article had the link to the code so don't even try to imply that I was hiding jack squat. He hasn't nailed shit, it's the same test epoll proponents have used for years. It's not a "localhost" test, it's a test that makes a bunch of file descriptors and then compares the poll vs. epoll performance as active vs. total changes.

But hey, you can live in your own little fantasy world where you think you've won some kind of battle of the HN because you listened to some epoll fanatic weirdo and cheered him on.