[Please use the mailing list] On 26/04/13 08:30 AM, Marcelino Suzuki wrote: > Hello Sebastien > > And sorry to send you so many messages, but I am still a bit > confused. In short > > 1. The reason I really want Ray to work is because the "biological > abundances" and taxonomy functions (I got some alternative > assemblies). I have had problems with submitting jobs (which crashed > right away), but I was finally able to run on 16 cores in 8 nodes > using the miniranks option, One thing I noticed (by checking the > cluster node usage, is that one of the nodes gets all the memory load > and cpu usage, and as I mentioned in a previous mail, all nodes output > Rank 0 and only one of the nodes node.1 gets the progress of the > output written onto it. I just want to make sure I am not somewhat > using just using one of the nodes in this run. Is the biological > abundances part of Ray not "parallel?" >
No. it is distributed across the allocated ranks. > I am sending below my command file > > job2x16k31bimini.cmd > > #!/bin/sh > # @ job_name = crambe_ray > # @ output = $(job_name).out > # @ error = $(job_name).err > > # @ job_type = mpich > # @ node =8 > # @ total_tasks = 16 > # @ restart = yes > > > # @ wall_clock_limit = 60:00:00,59:55:00 > # @ queue > > /opt/cluster/mpi/intel/openmpi-noll-1.4.1-qlc/bin/mpiexec -n 8 -output- > filename cr -machinefile $LOADL_HOSTFILE /work/OOBMECO/bin/ray/Ray > -mini-ranks-per-rank 1 > -read-write-checkpoints checkpoints > -k 31 > -amos The -amos option will do a lot of input/output operations. > -o crambe.3 > -s /scratch/suzukim/mira/CCB1a.fastq > -search /scratch/suzukim/NCBI-taxonomy/NCBI-Eukaryote-Genomes > -search /scratch/suzukim/NCBI-taxonomy/NCBI-Finished-Virus-Genomes > -search /scratch/suzukim/NCBI-taxonomy/NCBI-Draft-Bacteria-Genomes > -search /scratch/suzukim/NCBI-taxonomy/NCBI-Finished-Bacterial-Genomes > -with-taxonomy /scratch/suzukim/ggenes-taxonomy/Genome-to-Taxon.tsv / > scratch/suzukim/ggenes-taxonomy/TreeOfLife-Edges.tsv /scratch/suzukim/ > ggenes-taxonomy/Taxon-Names.tsv > -search /scratch/suzukim/Ontology/EMBL_CDS_Sequences > -gene-ontology /scratch/suzukim/Ontology/OntologyTerms.txt /scratch/ > suzukim/Ontology/Annotations.txt > -show-memory-usage > > Which version of Ray are you using ? What type of cluster are you using ? > = > = > = > = > ======================================================================== > oOOOOo Marcelino Suzuki, Professor, Team > Leader, Platform Responsible > oOOO Univ Pierre Marie Curie (Paris 6) - > Observatoire Océanologique de Banyuls > oOOOOOo. UMR 7621 - Laboratoire > d'Océanographie Microbienne (LOMIC) > .oOOOOOOOo. Marine Biodiversity and > Biotechnology (bio2mar) Platform > .oOOOOOOOOOoo. suz...@obs-banyuls.fr http://bit.ly/fq3nbE > bio2mar.obs-banyuls.fr > .oOOOOOOOOOOOooooo. Ave du Fontaulé, Banyuls-sur-Mer 66650, France > +33(0)430192401 > 0000000000000000000000000000000000000000000000000000000000000000000000000000 > > = > = > = > = > ======================================================================== > oOOOOo Marcelino Suzuki, Professor, Team > Leader, Platform Responsible > oOOO Univ Pierre Marie Curie (Paris 6) - > Observatoire Océanologique de Banyuls > oOOOOOo. UMR 7621 - Laboratoire > d'Océanographie Microbienne (LOMIC) > .oOOOOOOOo. Marine Biodiversity and > Biotechnology (bio2mar) Platform > .oOOOOOOOOOoo. suz...@obs-banyuls.fr http://bit.ly/fq3nbE > bio2mar.obs-banyuls.fr > .oOOOOOOOOOOOooooo. Ave du Fontaulé, Banyuls-sur-Mer 66650, France > +33(0)430192401 > 0000000000000000000000000000000000000000000000000000000000000000000000000000 > ------------------------------------------------------------------------------ Try New Relic Now & We'll Send You this Cool Shirt New Relic is the only SaaS-based application performance monitoring service that delivers powerful full stack analytics. Optimize and monitor your browser, app, & servers with just a few lines of code. Try New Relic and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr _______________________________________________ Denovoassembler-users mailing list Denovoassembler-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
