News for hackers. Stuff that matters.

A comment on The Case for a High-Level Kernel-Bypass I/O Abstraction. Parent comment

Re:The Case for a High-Level Kernel-Bypass I/O Abstraction by yourbandsucks on Thursday May 23, 2019 @08:21AM (#19990975)
So what's the API look like?

You said you don't want to make users deal with flow control and hardware details.. does that imply a userspace bypass library which does that stuff for us? Does it look posixy?

[ Parent | on HN ]
Re:The Case for a High-Level Kernel-Bypass I/O Abstraction by SEJeff on Thursday May 23, 2019 @10:05AM (#19991901)
Solarflare's OpenOnload or Mellanox's VMA both show up as LD_PRELOADS that overload any traditional socket programming unless you want to code your apps to their API directly.
[ Parent | on HN ]
Re:The Case for a High-Level Kernel-Bypass I/O Abstraction by iyzhang on Thursday May 23, 2019 @08:49AM (#19991182)
It looks POSIX-like but uses high-level queues and fixes some issues with epoll. The lack of an atomic data unit and the overhead of the poor epoll interface cost too much to retain for kernel-bypass. Take a look at the paper for more details.
[ Parent | on HN ]
Re:The Case for a High-Level Kernel-Bypass I/O Abstraction by amluto on Thursday May 23, 2019 @10:04AM (#19991883)
Where’s the paper? After looking at your site, it’s not obvious to me what paper to look at.
[ Parent | on HN ]
Re:The Case for a High-Level Kernel-Bypass I/O Abstraction by iyzhang on Thursday May 23, 2019 @10:15AM (#19991992)
Paper can be found here: http://irenezhang.net/papers/demikernel-hotos19.pdf
[ Parent | on HN ]