[Bioclusters] mpiBLAST

[Joe Landman]

On Wed, 2003-05-14 at 09:29, Osborne, John wrote:
> Hi Joe,
> 
> I wrote an email earlier for another thread where I described some of the
> problems I ran
> into.  In short I was complaining about:
> 
> 1)mpiformatdb creating 21 database fragments when I tell it to create 20, or
> 19 when I tell it to create 18... ie)n0 to n20 when I ask for just 20

Hi John:

  Aaron Darling and the other mpiBLAST folks are computer scientist
types, who count starting at 0.  So N fragments would be labeled from 0
.. N-1.  This should be more clear in the README.

> 2)The (null).pal file in my shared storage directory (created by
> mpiformatdb).  I think the
> file is also incorrect since it contains a line like this:
> DBLIST (null).00 (null).01 ... (null).20

The .pal is the list of DBs to use for the search, so the parameters get
the search space size correct.  If you have (null).pal, I think there
may have been a db naming error on the mpiformatdb line.  Could you
indicate what you used?

> 3)The misleading name of mpiformatdb which doesn't actually have an MPI
> library calls (I grepped
> for MPI_).  This is just a pet peeve because I was upset at how long the
> format took.  :)

The funny thing is that there is a copy phase, a format phase, and
another copy phase.  It is time consuming, but not too painful....

> Right now I have it running, but with it doesn't produce any output and
> lamclean hasn't killed
> the process on the first node (it has been trying for >5 minute now).
> Speaking of the first node,
> are we supposed to set up n0 as our master node?  Is that why I get 21
> fragments when I say divide
> it into 20 pieces?

Hmmm...  I have not built mpiBLAST with LAM (I use MPICH).  Do you have
MPICH available, or do you use only LAM?  If so, I would be curious
about how to make it work there.  I might set it up on my laptop so I
can experiment.

> 
>  -John
> 
> 
> 
> -----Original Message-----
> From: Joe Landman [mailto:landman@scalableinformatics.com]
> Sent: Tuesday, May 13, 2003 7:57 PM
> To: biocluster
> Subject: RE: [Bioclusters] mpiBLAST
> 
> 
> On Tue, 2003-05-13 at 17:27, Osborne, John wrote:
> > RedHat 7.2 with the 2.4.19 kernel (SMP)
> > 
> > I can get it to compile, but I have issues running it.
> 
> If you are stuck trying to run it, please let us know what the errors
> you are getting are.
> 
> Would it help if I wrote a more detailed step-by-step, complete with
> examples?
-- 
Joseph Landman, Ph.D
Scalable Informatics LLC,
email: landman@scalableinformatics.com
web  : http://scalableinformatics.com
phone: +1 734 612 4615