Yes.  I wasn't clear.  I apologize for that.  Mostly I was referring to
hardware-level system design, and how it affects design and performance
of a piece of software.  Or how the design and operation of a piece of
software has an effect on the system as a whole.

For example, at my work we often have to explain to our users (not
exclusive CS people, but they're certainly included), what I'd consider
basic concepts like why striding outside the processor cache too
frequently is killing their performance, or why its better to have
fewer, larger files on the storage system instead of many million that
are tiny, or why storing your DNA sequence data in an uncompressed ASCII
format (they often generate 10's of TB per day per user) is a bad idea,
especially when there's a compressed format easily available (with
frequently 5:1 compression ratios).

I can be more specific if you all want, but it might take me a little
bit to make sure I don't turn it into a "stupid user" rant.  They're
good at what they do and everything.  It's just surprising sometimes
when you see the gaps in their understanding, and have to go back and
explain.



Lloyd Brown
Systems Administrator
Fulton Supercomputing Lab
Brigham Young University
http://marylou.byu.edu

On 08/06/2013 10:03 AM, David Hilton wrote:
>     - Computer Science - Lots of algorithms and programming, but absolutely
>     nothing on systems or hardware
> 
> 
> It depends on what you mean by systems.
--------------------
BYU Unix Users Group
http://uug.byu.edu/

The opinions expressed in this message are the responsibility of their
author.  They are not endorsed by BYU, the BYU CS Department or BYU-UUG.
___________________________________________________________________
List Info (unsubscribe here): http://uug.byu.edu/mailman/listinfo/uug-list

Reply via email to