Re: quotas [was: thread ids, task ids and subsystems]

[Michal 'hramrach' Suchanek]

On Wed, Apr 16, 2003 at 01:17:20PM +0200, Marcus Brinkmann wrote:
> On Wed, Apr 16, 2003 at 12:01:08PM +0200, Michal 'hramrach' Suchanek wrote:
> > IIRC most quota problems were already discussed specifically for memory.
> 
> I am unaware of any discussion of these issues.
That is the "currency based virtual memory" which looks to me like memory
quotas by a different name.
> 
> > The protocol for obtaining memory could be probably generealized for any
> > resource.
> 
> i think you still didn't take into account the unique design features of the
> Hurd.
> 
> > I cannot remember any solution for the problem of reclaiming resources, 
> > which
> > is probably the same for all types of resources.
> 
> If I can not reclaim a resource I provided to an untrusted server, then the
> design is incomplete.

Indeed.
> 
> > It looks like there are basically two types of resources that a user could
> > spend:
> >  - real hardware resources (like memory, cpu time, network bandwidth,
> >     disk space, ...)
> 
> Those are the only resources that matter.
> 
> >  - some artifical resources that may correspond to system resources 
> > indirectly
> > (ie database transactions, file handles, sound records)
> 
> There is no need to care about these in the Hurd.  A file handle is just a
> connection to a server and should be accounted for with the resources the
> server needs to keep the connection.

But it's the user's option to give his raw system resources to a server
process that uses a different type of accounting (ie database transactions).
There's just the question if this could be integrated seamlessly with the
accounting of raw system resources. A general interface could accomplish
this and it would come handy in case new 'real' resources are identified
(or the possibility to impose restrictions on them implemented).

--
Michal Suchanek
address@hidden