Re: ATS QUIC Hackathon Result

> On Nov 19, 2018, at 19:14, 宋辰伟 <616955...@qq.com> wrote: > > Currently QUIC event doesn’t use the epoll in ET_NET, SO it is a really a > problem. Ah. Yes, part of our planning session included moving all this to ET_NET. That is part of using connect+bind on the QUIC sessions. Cheers, — L

Re: ATS QUIC Hackathon Result

Currently QUIC event doesn’t use the epoll in ET_NET, SO it is a really a problem. > 在 2018年11月20日，上午10:10，Leif Hedstrom 写道： > > >> On Nov 19, 2018, at 18:47, 宋辰伟 <616955...@qq.com> wrote: >> >> Hi >> >> Actually there is some blocking operation during ats like epoll. We are >> blocking a

Re: ATS QUIC Hackathon Result

> On Nov 19, 2018, at 18:47, 宋辰伟 <616955...@qq.com> wrote: > > Hi > > Actually there is some blocking operation during ats like epoll. We are > blocking at epoll for several times and until the net event arrived. So if > any other event (which is not net event) schedule it will be pended u

Re: ATS QUIC Hackathon Result

Hi masaori Sorry for the wrong name. Hi maskit and masaori Fill an issue try to figure out the receive side delays . scw00 > 在 2018年11月20日，上午9:21，Walt Karas > 写道： > > What strategy or strategies do we use in ATS to make sure that we > don't do blocking I/O

Re: ATS QUIC Hackathon Result

Hi Actually there is some blocking operation during ats like epoll. We are blocking at epoll for several times and until the net event arrived. So if any other event (which is not net event) schedule it will be pended until epoll’s timeout fired unless you use the signal to terminate blocking

Re: ATS QUIC Hackathon Result

All network IO is non blocking, disk IO is scheduled on dedicated IO threads (scales by number of disks). The latter is essentially our own AIO, but we also support Linux native AIO. — Leif > On Nov 19, 2018, at 18:21, Walt Karas wrote: > > What strategy or strategies do we use in ATS to mak

Re: ATS QUIC Hackathon Result

What strategy or strategies do we use in ATS to make sure that we don't do blocking I/O that blocks a thread with queued event handlers (not dependent on the I/O operation)? On Mon, Nov 19, 2018 at 7:13 PM Masakazu Kitajo wrote: > > I have no idea where the delay comes from. I posted its detail on

Re: ATS QUIC Hackathon Result

I have no idea where the delay comes from. I posted its detail on the PR ( https://github.com/apache/trafficserver/issues/3552#issuecomment-440099449). Thanks, Masakazu On Mon, Nov 19, 2018 at 5:30 PM 宋辰伟 <616955...@qq.com> wrote: > Hi maskit and koshiba > >What does 30ms delay means ? How t

Re: ATS QUIC Hackathon Result

Hi maskit and koshiba What does 30ms delay means ? How to reproduce it ? SCW00 > 在 2018年11月19日，上午9:12，Masakazu Kitajo 写道： > > seem