Re: [2.4.17/18pre] VM and swap - it's really unusable

[Daniel Phillips]

On January 16, 2002 04:27 pm, Bill Davidsen wrote:
> On Tue, 15 Jan 2002, Daniel Phillips wrote:
> > On January 15, 2002 06:26 am, Mark Hahn wrote:
> > > > than the task's float, the completion time of the schedule as a whole will be 
> > > > delayed.  This is no different for a computer than it is for a group of 
> > > > people, it is still a scheduling problem.  Delaying any random task risks 
> > > 
> > > it is quite different.  with computers, there are often STRONG benefits
> > > to clustering, batching, chunking, piggybacking, whatever you want to call it.
> > 
> > It's no different.
> 
> Sorry, there are strong benefits from all of the things mentioned. I lack
> time and inclination to explain how caching works, but there are costs of
> changing from one thing to another.

With people, there are often STRONG benefits to clustering, batching, 
chunking, piggybacking.  See what I mean?

> The other issue is that processes doing i/o (blocking before a whole
> timeslice) will run better if they get priority when they can use the CPU.
> Therefore a system needs to recognize (and be tuned) for both of these.
> 
> Computers are very different than people in lines.

Oh yes, computers don't *know* they are in lines.

--
Daniel