🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.sys.super
Article #393

Re: What are people paying for when they by a supercomputer?

#393
From: Eugene N. Miya
Date: Thu, 19 Nov 1992 18:28
63 lines
3091 bytes
In article <6080@mccuts.uts.mcc.ac.uk> zzassgl@uts.mcc.ac.uk
(Geoff Lane) writes:
>MIPs aren't the problem - I/O and memory access is the problem.  Any old 
>fool can plug together a bunch of Intel CPU chips and claim that the 
>resulting box is faster than a Cray (usually the model isn't specified :-) 
>but it takes an expert to get the bus architecture correct.  It's also damn 
>expensive.

Sort of.
Some would say you buy "balance."  My old officemate George came up with this
nice concept of how you think of this:

                   ^ A  
                s / \ r                
               m /   \ c              
              h /     \ h            
             t /       \ i          
            i /         \ t        
           r /           \ e      
          o /             \ c    
         g /               \ t  
        l /                 \ u
       A /                   \ r
        <_____________________> e   
                Language

It is the icon for the Conferences on High-Speed Computing
which has some characteristics to Amdahl's Other Law (remote).
Gets you out of binary thinking.
There is an argument in another group and mailing list right now about
a slight against software inproving performance.
We hash on 3-D versions, but here is my "improved" 2-D version:

                         ^  
                        /.\  
                       / . \  
                    e /  . O\ S
                     /   .  p\  
                  r /n   .C  e\ o
                   /o   C.o   r\  
                a /i     .m    a\ f
                 /t     P.p     t\  
              w /a       .i      i\ t
               /c       U.l       n\ 
            d /i         .e        g\ w
             /n          .r          \ 
          r /u           .           S\ a
           /m           y.L           y\
        a /m         r.  N  .i         s\ r
         /o       o.  a     o  .b       t\
      H /C     m.  t           t  .r     e\ e
       /    e.  a                 a  .a   m\
      /  M.  D                       t  .r  \
     / .              D a t a           i  . \
    <_________________________________________> 
               A p p l i c a t i o n

--eugene miya, NASA Ames Research Center, eugene@orville.nas.nasa.gov
  Associate Editor, Software and Publication Reviews
  Scientific Programming
  {uunet,mailrus,other gateways}!ames!eugene
Seeking Books to buy:	Bongard, Pattern Recognition
			3 down 1 to go.
Technical Details
Message-ID: <1992Nov19.182855.8284@nas.nasa.gov>
Path: arc1-ca.pugleaf.net!sparky!uunet!think.com!ames!data.nas.nasa.gov!wilbur.nas.nasa.gov!eugene
References: <1992Nov16.141214.7632@walter.bellcore.com> <1992Nov16.174414.12427@craycos.com> <6080@mccuts.uts.mcc.ac.uk>

Show Full Headers
Complete Article Headers
Newsgroups: comp.sys.super
Path: arc1-ca.pugleaf.net!sparky!uunet!think.com!ames!data.nas.nasa.gov!wilbur.nas.nasa.gov!eugene
From: eugene@wilbur.nas.nasa.gov (Eugene N. Miya)
Subject: Re: What are people paying for when they by a supercomputer?
References: <1992Nov16.141214.7632@walter.bellcore.com> <1992Nov16.174414.12427@craycos.com> <6080@mccuts.uts.mcc.ac.uk>
Sender: news@nas.nasa.gov (News Administrator)
Organization: NAS, NASA Ames Research Center, Moffett Field, CA
Date: Thu, 19 Nov 92 18:28:55 GMT
Message-ID: <1992Nov19.182855.8284@nas.nasa.gov>
Lines: 63