Re: aio_read implementaion in Squid

Hello joe,

I was asking this for Net I/O. I had a cursery look on eventio and signal
driven io branches. eventio branch also was done by Henrik and adrian (If I
am right). I already asked adrian about eventio, but i did not get any
reply. It seems he is busy with his work now. I got it what you are telling.
But i want to know if we go to aio for the Net I/O, what are the issues we
have to face?

I do not know you are remembering or not. I had asked you many Qs about CPU
Usage. Now my objective is change the poll in Squid. I am looking for very
high performance Squid. I want to get feedbacks from you all before i
proceed. And I am very curious to know about the OS compatibilities problem,
If we go for aio.

Regards,
Venkatesh

> Adrian has implemented an aio interface to COSS.
>
> Might be worth a look if you need pointers for how to proceed (though
> COSS is significantly different than UFS).
>
> I doubt you'll find it provides any higher performance than the current
> AUFS implementation (but then again, AUFS only works on Linux and
Solaris).
>
> Venkatesh wrote:
>
> > Hi All,
> >
> > What are the issues if we go for aio_read() instead of read() with
> > poll() other than posix compatibility?
> >
> > Do you all agree with this kind of implementaion for Squid?
> >
> > Any comments?
> >
> > Venkatesh
>
>
> --
> Joe Cooper <joe@swelltech.com>
> Affordable Web Caching Proxy Appliances
> http://www.swelltech.com

_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com

Received on Wed Sep 12 2001 - 00:07:46 MDT