IUBio

GCG under Linux

Chuck Dillon cdillon at nimblegen.com
Fri Sep 5 08:52:23 EST 2003


Steve Thompson wrote:
> ...   I agree that the 32-bit architecture is a restriction and
> GCG would undoubtedly run faster on a 64-bit system.  

The vast majority of the processing in the Wisconsin Package is 
character based and as such is not going to benefit from 64bit 
addressing or arithmetic.  Assuming the problem fits in a 32 bit 
address space the best you can expect is parity on a 64bit box with 
comparable processing speed as a 32bit box.  Almost certainly the 64bit 
box will be somewhat slower because of the overhead of 64bit hardware 
and larger executables.

This is true in general when comparing 32 to 64 bit.  The advantage of 
64bit is the larger address space and performance when doing 
computations that can't be done intrinsically in 32bits.  Generally 
speaking.

-- ced




-- 
Chuck Dillon
Senior Software Engineer
NimbleGen Systems Inc.




More information about the Info-gcg mailing list

Send comments to us at biosci-help [At] net.bio.net