[Pw_forum] Pentium-M versus Itanium

[Miguel Martínez Canales]

Dear all,

Not wanting to lose all fun from CVS espresso, and after running unstable 
linux distros for a while, I decided to finally give the CVS version a go. 
I then compiled it in both my laptop and in the entry node of my 
university's cluster. Specs follow:

Laptop:
  Pentium-M 735 (1.7 GHz), 512 Mb RAM, running Ubuntu Hardy (kernel 2.6.24, 
gcc and gfortran 4.2.3)

PC:
  Pentium4 (2.6 GHz), 512 Mb RAM, running Debian testing (kernel 2.6.22, 
gcc and gfortran 4.3.0 rc2

Big Bad Machine:
  2*Itanium 2 @ 1.3GHz, 2Gb RAM, running RHEL ¿4? (kernel 2.6.9, ifort 10 
and mkl 10)

The itanium build is probably hampered by a parallel build, but Ubuntu and 
Debian run the script with bash (because dash doesn't work with 
check_failure), and with the supposed disadvantage of gcc. I've also needed 
to do a couple of hops to compile it on the Debian based machines. In any 
case, after loading a couple examples into memory (yeah, I know, not very 
stressful), I get:

Laptop:
   example01
real	1m44.686s
user	1m41.374s
sys	0m1.864s
   example05
real	0m8.411s
user	0m7.828s
sys	0m0.140s

PC
   example01
real	1m46.421s
user	1m41.742s
sys	0m3.632s
   example05
real    0m7.505s
user    0m7.184s
sys     0m0.260s

Big Bad Itanium
   example01
real	1m48.417s
user	2m15.136s
sys	0m6.756s
   example05
real	0m13.028s
user	0m12.775s
sys	0m0.627s

Is it normal that user time > real time even when I am running pw.x as 
mpirun -np 1 pw.x? Is it normal that my almost 4 year old laptop running a 
bloated distro is running the examples clearly faster than an Itanium 
machine? I swear I haven't used fancy flags anywhere, other than the 
default ones. I will also run espresso on an opteron machine, to see the 
results.

Regards,

Miguel

PS: I also swear that the very same Itanium machine used to run twice as 
fast as the same laptop, although the free FORTRAN compilers have changed a 
lot in the last couple of years.

-- 
----------------------------------------
Miguel Martínez Canales
    Dto. Física de la Materia Condensada
    UPV/EHU
    Facultad de Ciencia y Tecnología
    Apdo. 644
    48080 Bilbao (Spain)
Fax:  +34 94 601 3500
Tlf:  +34 94 601 5326
----------------------------------------

  "If you have an apple and I have an apple and
  we exchange these apples then you and I will
  still each have one apple. But if you have an
  idea and I have an idea and we exchange these
  ideas, then each of us will have two ideas."

  George Bernard Shaw