On 30-Mar-98 Ty Sarna wrote:

> In article <XFMail.980329044125.jiho@mail.c-zone.net> you write:
>> I do hope UVM is more efficient that this sucker.  I must say, though, I have
>
> [I know you were talk memory efficiency,
> but I just had to rave for a second]
>
> At least speedwise, between the new (old) vfork(), UVM, and PMAP_NEW,
> -current feels a _LOT_ faster doing things like "make build". Program
> start-up time seems to have been drastically cut. 

Now that raises an interesting point.  There's more than one way to look at
speed.  Nothing slows a system down like going into swap.

In the past, I've noticed a lot of attention being paid to improving the way
systems handle memory shortages, but not much to minimizing their occurence to
begin with.  As far as I can tell, any improvement in the latter has been a
side-effect of work on the former.

Programmers have long made the argument, well, memory will be so cheap, we can
optimize for speed at the expense of space.  So the system runs very fast for a
little while, until it starts thrashing swap.  Soon it stops running entirely. 
So you wind up with a system that is very fast at becoming unusable for any
serious purpose.

Meanwhile, CPUs have gotten so fast, you could have argued the exact reverse. 
And much of the "optimization" that costs space turns out to be sheer
sloppiness anyway, unrelated to any real speed gains except the programming
time saved by not paying attention.

This is NOT to belittle the vm work Chuck and others have done, which certainly
seems to fill a need -- I just haven't had a chance to evaluate any of it yet,
so I don't know if my concerns are answered (to the extent that my concerns
involve the way the kernel itself works).  The desire to do so is obstructed by
the necessity of installing a complete new distribution, and setting up a
working envoirnment almost from scratch.


--Jim Howard  <jiho@mail.c-zone.net>


----------------------------------
E-Mail: jiho@mail.c-zone.net
Date: 30-Mar-98
Time: 16:32:00

This message was sent by XFMail
----------------------------------