Re: MINA read/write starvation

[Trustin Lee <tr...@apache.org>]

On 9/29/06, mark.atwell@jpmorgan.com <ma...@jpmorgan.com> wrote:
>
> Hi Trustin,
>
> You may recall  I contributed some fixes and enhancements to MINA many
> months ago.
>
> I seem to recall that we (or possibly you and one of my colleagues -
> Robert
> Greig) may have even exchanged e-mails - or possibly some Jiras in the
> past
> about this:
>
> I think we encountered some read starvation problems when we had an app
> that was generating large amounts of writes (though it could be
> bi-directional).
>
> Unfortunately I don't recall the conclusion(s). Is this now 'fixed'? I
> trawled the MINA Jira and found some similar sounding issues e.g.
> DIRMINA-206 and 262 but can't really conclude anything.


DIRMINA-206 resolves read starvation.  For write starvation, we are still
thinking about it, but it basically can be resolved by implementors because
implementors can control it.  You can monitor the remaining number of bytes
in the write queue by calling IoSession.getQueuesWriteBytes().  (Sorry. I
don't remember the exact name. ;)

Of course, we want to keep improving starvation prevention.  Any feedback or
idea are welcome!

Trustin
-- 
what we call human nature is actually human habit
--
http://gleamynode.net/
--
PGP key fingerprints:
* E167 E6AF E73A CBCE EE41  4A29 544D DE48 FE95 4E7E
* B693 628E 6047 4F8F CFA4  455E 1C62 A7DC 0255 ECA6