Archives


- Beowulf
- Beowulf Announce
- Scyld-users
- Beowulf on Debian

[Beowulf] Bioinformatics Benchmark System v3 rc1 ready

Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.

Search

Jan-Frode Myklebust janfrode at parallab.uib.no
Mon Jul 12 06:31:45 PDT 2004


On Mon, Jul 12, 2004 at 08:55:44AM -0400, Joe Landman wrote:
> 
> On the Perl modules, could you let me know which ones were not in the
> RHEL 3 AS on PPC?  I included a method to handle missing modules, though
> I don't know what this particular version of RHEL requires.  

When doing 'make' under perl_modules I get:

cd   Getopt-Long-2.34_01 ; perl Makefile.PL ; make ; make test ; make install
 
WARNING: This is Getopt::Long version 2.34_01.
The underscore in the version number indicates that this is
an intermediate, preview or beta release.
Please keep checking CPAN for the official release.
 
 
Warning: I could not locate your pod2man program. Please make sure,
         your pod2man program is in your PATH before you execute 'make'
 
Writing Makefile for Getopt::Long
make[1]: Entering directory `/scratch/janfrode/bbsv3rc1/perl_modules/Getopt-Long-2.34_01'
Makefile:87: *** missing separator.  Stop.
make[1]: Leaving directory `/scratch/janfrode/bbsv3rc1/perl_modules/Getopt-Long-2.34_01'
make[1]: Entering directory `/scratch/janfrode/bbsv3rc1/perl_modules/Getopt-Long-2.34_01'
Makefile:87: *** missing separator.  Stop.
make[1]: Leaving directory `/scratch/janfrode/bbsv3rc1/perl_modules/Getopt-Long-2.34_01'
make[1]: Entering directory `/scratch/janfrode/bbsv3rc1/perl_modules/Getopt-Long-2.34_01'
Makefile:87: *** missing separator.  Stop.
make[1]: Leaving directory `/scratch/janfrode/bbsv3rc1/perl_modules/Getopt-Long-2.34_01'
make: *** [Getopt-Long] Error 2

% which pod2man
/usr/bin/pod2man
% rpm -qf /usr/bin/pod2man
perl-5.8.0-88.4



When trying to run 'bbsrun' I get:

% ./bbsrun
Can't locate Object/MultiType.pm in @INC (@INC contains: /usr/lib/perl5/5.8.0/ppc-linux-thread-multi /usr/lib/perl5/5.8.0 /usr/lib/perl5/site_perl/5.8.0/ppc-linux-thread-multi /usr/lib/perl5/site_perl/5.8.0 /usr/lib/perl5/site_perl /usr/lib/perl5/vendor_perl/5.8.0/ppc-linux-thread-multi /usr/lib/perl5/vendor_perl/5.8.0 /usr/lib/perl5/vendor_perl /usr/lib/perl5/5.8.0/ppc-linux-thread-multi /usr/lib/perl5/5.8.0 .) at /usr/lib/perl5/site_perl/5.8.0/XML/Smart.pm line 18.
BEGIN failed--compilation aborted at /usr/lib/perl5/site_perl/5.8.0/XML/Smart.pm line 18.
Compilation failed in require at ./bbsrun line 76.
BEGIN failed--compilation aborted at ./bbsrun line 76.


> As for the fancy XML input and output, one of the intended uses of this
> tool is for automatic generation of benchmark tests and results for
> complex distributed systems in conjunction with other tools under
> development, which might not always have a bourne shell available. 
> Additionally, the input system is meant to be as flexible and simple as
> possible, so we try to avoid being "fancy" :)

Then maybe you should consider splitting it into separate programs? Like
bonnie++ which does simple comma-separated output (csv), which can be fed
to other tools for fancy-formating.

The fact that it failed to run for me here is a bit of a turn off, and
it's not very tempting to try to fix it on every platform I want to use in
my current benchmarking (Currently AIX, RedHat 7.1, RHEL 3 on x86, RHEL 3
on ppc970 and Rocks).  Then it's much easier to pick the simple command
lines from baseline/*xml and execute them by hand.  

BTW: thanks a lot for providing these benchmark configurations. They're
very usefull to us! 


> The bbsv1 benchmarks?  These are the original from the previous
> version.  The baselines are (eventually) supposed to cover the small
> through huge cases.  

Yes, I've only run blast from the previous version. And it seemed to be
very much affected by the 1 GB memory on our nodes. 

> What I didn't mention in the docs was database formatting.  You can use
> formatdb -v N  and set N large enough such that the indices do not
> overflow buffer cache.  One of the uses of this tool is to figure out
> where this size might be, by examining the time of BLAST execution
> versus database volume size.  I will try to work up an example of this
> in the next few days and get this onto the page.

OK, great!


  -jf



More information about the Beowulf mailing list