Re: SV: suggestions for udb

["Wes Bauske" <wsb paralleldata com>]

Shane Sturrock wrote:
> 
> 
> Micro-parallelism which uses the 64 bit word to do multiple integer
> operations on several less than 64 bit numbers.  By using this we can do
> the full Smith and Waterman algorithm at the sort of speeds others are
> getting from the cut down heuristic algorithms and not suffer the
> sensitivity penalties that they do.  It's really neat stuff!
> 
> Here is a reference that covers the topic pretty thoroughly:
> 
> http://www.supercomp.org/sc95/proceedings/549_LCAR/SC95.HTM
> 
> This was the work that got us interested.
> 

Interesting. Similar to what graphics folks do. Given
that sort of implementation, the Intel should actually
run very fast since it has those MMX instructions for
just that sort of problem. It would be similar to what
the paper talks about for the I860. You could build a
simple asm macro instead of what the authors do for the 
i860.

Not sure if I'd bother though unless it was really
fast and #ifdef'd with equivalent C.

Still, I see no reason why one couldn't put two 15 bit words
in a 32 bit word on an intel and improve it's performance
as is done with 4 15 bit words on Alpha. Also, one could 
implement the FP version of the algorithm and cut the
Alpha's lead that way too. 

One question though. Why are you comparing only 15 bits
at a time? Is this something intrinsic to the problem?
Why not directly use 32 or 64 bits??


Wes