l4-hurd
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Resource Management on General-Purpose Systems


From: Marcus Brinkmann
Subject: Re: Resource Management on General-Purpose Systems
Date: Thu, 12 Jul 2007 19:26:39 +0200
User-agent: Wanderlust/2.14.0 (Africa) SEMI/1.14.6 (Maruoka) FLIM/1.14.8 (Shijō) APEL/10.6 Emacs/23.0.0 (i486-pc-linux-gnu) MULE/6.0 (HANACHIRUSATO)

At Thu, 12 Jul 2007 16:41:59 +0200,
Neal H. Walfield wrote:
> I disagree with his claim that there is a near abundance of resources.
> First, the data that is manipulated on these systems is growing in
> size and complexity.  The resolution of digital photos and movies are
> increasing and web pages contain rich content such as Flash or Ajax
> applications.

It seems to me it is worthwhile to make a finer distinction here.
Flash and Ajax stay hot in core most of the time (I just killed off a
few tabs in my browser because I had 24% CPU utilization for
advertisement banners).  Ie, they stay in the active working set.

Digital movies are streamed, and the active working set is only a
small part at any time.  Jonathan makes a point about video editing,
but as the files just don't fit into core no matter how you look at
it, you really have no choice but to edit frame by frame (I am not
convinced the problem is mmap() here, but then, maybe I misunderstood
something).

As for pictures, high resolution is a bit of a problem.  The pic can
be downscaled, but of course you want fast and seamless scaling.  I
was quite impressed by the Photosynth presentation by Microsoft [1]
which acquired Seadragon [2] recently.  Seadragon they claim is not
bound by data size, but by ratio of bandwidth to pixels on the screen.
However, I do not know how they do it.  Anybody?

[1] http://www.ted.com/index.php/talks/view/id/129
    Health Warning: The advertisement before the talk is ear-damaging loud.
    Start at second 25.
[2] http://labs.live.com/Seadragon.aspx





reply via email to

[Prev in Thread] Current Thread [Next in Thread]