[frognibble]

The blog post does not say if the epoll code uses level triggering or edge triggering. It would be interesting to see the results for both modes. The smaller number of system calls required for edge triggering might make a difference in performance.

[zedshaw]

That's entirely possible, but then you pay a penalty in complexity because you have to keep track of missed events yourself. I think (unproven) that it's actually a wash because of this.

[frognibble]

At most, you need to track a couple of booleans per socket, one for read and one for write.

Depending on what you are doing, you might not even need to track these booleans. For example, on the read side you can ignore read events when you are not interested in reading. When you switch back to read interest, you can read the socket to see if data arrived while you ignored events. A similar strategy can be used on the write side.