[BioC] aCGH package - performance Issues

Ramon Diaz-Uriarte rdiaz at cnio.es
Wed May 7 11:35:33 CEST 2008

Dear Julian,

We have not heard of any reports (or success or lack of it) on a Solaris 
machine. Our code does depend heavily on Rmpi (and papply). Rmpi itself will 
run with both OpenMPI and LAM/MPI. But, if I understand correctly, either one 
should work OK in Solaris. If either OpenMPI or LAM/MPI can be installed and 
made to work on your machine, then I assume everything else should work just 

From the point of view of the code, it does not matter if we are running over 
a cluster or on an SMP machine; once the MPI universe or equivalent is up and 
working, the load will be distributed to the slaves (which can live in 
different machines or in the same machine). With your setup, I'd do a couple 
of trial runs with either 4, 6, or 8 slave R processes.

Please let me know if you find problems or difficulties.



On Wednesday 07 May 2008 04:08, Julian Lee wrote:
> Hi R,
> wonderful article. 8 algorithms in MPI.
> I would so love to test your code but unfortunately i do not have the
> luxury of a linux cluster here(that however can be fixed ;)).
> I do however have a Sun v490, 4 dual core UltraSparcIV++ with 32GB RAM.
> I presume it'll work on this SMP too, however any concerns if i were to
> take this onto a Solaris machine?
> regards
> ----- Original Message -----
> From: "Diaz.Ramon" <rdiaz at cnio.es>
> To: "Julian Lee" <julian at omniarray.com>, "bioconductor"
> <bioconductor at stat.math.ethz.ch> Sent: Tuesday, May 6, 2008 2:19:36 AM GMT
> -08:00 US/Canada Pacific Subject: RE: [BioC] aCGH package - performance
> Issues
> Dear Julian,
> We have parallelized (over arrays or arrays * chromosomes) the calls to
> find.hmm (as well as other aCGH methods) using MPI. The R code is available
> from the ADaCGH package from CRAN. (The paper describing the approach,
> showing benchmarks, etc, is available from
> http://www.plosone.org/article/fetchArticle.action?articleURI=info%3Adoi%2F
> HTH,
> R.
> -----Original Message-----
> From: bioconductor-bounces at stat.math.ethz.ch on behalf of Julian Lee
> Sent: Tue 06-May-08 11:03
> To: bioconductor
> Subject: [BioC] aCGH package - performance Issues
> Hi all,
> I would like to know if there's a way to tweak the performance of the aCGH
> package, particularly the find.hmm.states function
> Input dataset
> Agilent CNV
> 31 samples
> 200,000 clones
> Hardware
> 2 Intel Xeon Dual Core 3GHz (total of 4CPUs)
> 4 GB RAM
> Windows 2003 Server Edition
> Software
> R version 2.7.0 (2008-04-22)
> i386-pc-mingw32
> locale:
> LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
> States.1252;LC_MONETARY=English_United
> States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
> attached base packages:
> [1] tools     splines   stats     graphics  grDevices utils     datasets
> [8] methods   base
> other attached packages:
> [1] aCGH_1.14.0     sma_0.5.15      multtest_1.20.0 Biobase_2.0.0
> [5] survival_2.34-1 cluster_1.11.10
> Function Call
> hmm(ex.acgh)<-find.hmm.states(ex.acgh)
> I am familiar with OpenMP. Is it possible to include these openMP pragmas
> into the function to speed up the computation? This is a concern as i will
> be moving onto an Illumina SNP dataset with 59 samples and 400,000 clones.
> Or would running it on a Linux machine be faster?
> dear moderators, Please direct me to the right forum if you think that this
> should be on the BioC-Dev mailing list instead.
> regards
> thank you
> --
> Julian Lee
> Bioinformatics Specialist
> Cellular and Molecular Research
> National Cancer Center Singapore
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
> **NOTA DE CONFIDENCIALIDAD** Este correo electrónico, y en su caso los
> ficheros adjuntos, pueden contener información protegida para el uso
> exclusivo de su destinatario. Se prohíbe la distribución, reproducción o
> cualquier otro tipo de transmisión por parte de otra persona que no sea el
> destinatario. Si usted recibe por error este correo, se ruega comunicarlo
> al remitente y borrar el mensaje recibido. **CONFIDENTIALITY NOTICE** This
> email communication and any attachments may contain confidential and
> privileged information for the sole use of the designated recipient named
> above. Distribution, reproduction or any other use of this transmission by
> any party other than the intended recipient is prohibited. If you are not
> the intended recipient please contact the sender and delete all copies.

Ramón Díaz-Uriarte
Statistical Computing Team
Centro Nacional de Investigaciones Oncológicas (CNIO)
(Spanish National Cancer Center)
Melchor Fernández Almagro, 3
28029 Madrid (Spain)
Fax: +-34-91-224-6972
Phone: +-34-91-224-6900

PGP KeyID: 0xE89B3462

**NOTA DE CONFIDENCIALIDAD** Este correo electrónico, y ...{{dropped:3}}

More information about the Bioconductor mailing list