lkml.org 
[lkml]   [2000]   [Feb]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: Scheduled Transfer Protocol on Linux
From
Date
> So, your position is "Linux isn't light enough to do the embedded job so
> let's bloat it up so we can handle the 64 processor case". Let's see.
> Millions of disk drives, millions of laptops, millions of small devices.
> How many 64 processor systems installed the world? Come again?

I'm saying it's already a bit bloated for low latency embedded systems,
compared to a special purpose system designed specifically for that task.

> The problem with your position is that given the fact that there are
> years of threading experience behind us, the only answer which makes
> sense is that the people who came before you didn't have the talent to
> get their locking model right. You keep asserting that it can be done
> and that sort of looks like you think you are smarter than the collective
> experience of the last 20 years of OS engineers.

I am not saying anything like that. The fact is the people before did get the
locking model right for their time. Hardware growth and more stress on
systems progressively bring out new bottlenecks. It takes years to identify
and fix those bottlenecks, and many of them can't be foreseen. Hardware
growth trends certainly outpace software growth trends, otherwise there
wouldn't be any problem with OS scaling.

> As much as I would love to think that you are that smart, statistics
> are heavily against it. If we can postulate that you are as good as the
> best who have come before you, but no better, then it seems a foregone
> conclusion that letting you go to town on the locking problem would
> result in another IRIX or Solaris.

Well what is best for certain classes of machines isn't better for others.
That will probably always be true. I take it that your opinion is that Linux
should target smaller systems. Personally, I think Linux has a great chance
of being a big player in the server market, but right now there is a bit of
catchup to be played to get Linux scaling excellently on this class of
machines.

> Well, you have chutzpah, I'll give you that.

Well thanks. Basically, I'm saying is that Linux seems to be at an indecisive
stage right now. It isn't clear to me whether it should proceed towards
embedded systems, single CPU machines, or midrange servers. I don't know of
any OS that works well across all of these platforms. Each has different
design goals and thus different philosophies that don't mesh well in the same
source base. Yet, people are working on getting Linux running well on each of
these targets and there seems to have been no real discussion about what kind
of source integration issues that will present.

The last thing I would want to see would be for Linux fall apart because of
various forces pulling it in different directions.

I may have made several rash points or given poor examples, but this is my
point in a nutshell.
--
Zach Amsden


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:56    [from the cache]
©2003-2011 Jasper Spaans