Multiple threads on ubuntu server
I am using Ubuntu 16.04 server. 88 threads, mem500Gb.
I used a program that supports multi-threads (number of threads=10)
In my case, it didn't use 10 CPUs. It is divided into a total of 10 processes and uses only 1 CPU.
I need any helps,,,
ubuntu
• 2,083 views
•
link
updated
by
colindaven
•
written
by
oghzzang •
0 answers
No answers yet.
Log in to answer this question.
More posts like this
-
Human sequence data mapped to mouse genome
written by sh •Hello everyone! Our lab now has a batch of single cell full length sequencing data - FLASH-Seq, - Paired-end, 151bp - Library Protocol: Nextera XT …
-
Tutorial: Single-cell RNA-seq: Preprocessing: Data integration and batch correction-3
written by Julia Ma •Part-1 here: https://www.biostars.org/p/9584982/ Part-2 here: https://www.biostars.org/p/9584985/ Full article lifted from: https://omicverse.readthedocs.io/en/latest/Tutorials-single/t_single_batch/ scVI ---- An important task of single-cell analysis is the integration of several samples, …
-
BLAST severe RAM throttling issues on Linux (Ubuntu)
written by Wilford203 •I have developed a Python based bioinformatics pipeline which utilises the BLASTx feature to BLAST a DNA sequence against a local copy of the nr …
-
CUDA error when I run megalodon (2.4.1) and guppy (6.0.1+652ffd1) with GPU support.
written by 1595952559 •Hello, everyone, recently when I run the megalodon command, there was a constant error. attached is my demo in the script. megalodon \ ~/01.data/01.ONT_data/02.ONT_test/AW/FAST5_PASS \ …
-
Command line BLAST is very slow
written by eager_underdog •Hi Fellow Bioinformaticians, I'm running command line BLAST on an EC2 instance (r4.8xlarg, 32 cpus and 244 GB memory), and it is super slow. I …
-
hisat2-build were killed when attempting to index the Ginkgo genome
written by zoukai3412085 •The Ginkgo genome is about 9.8GB, and I tried to build index in hisat2-build with the parameters --ss and --exon however, two times when I …
-
How should I deal with my RNAseq result with high % of mutiple aligned reads?
written by Simon Ahn •![enter image description here][1] [1]: /media/images/112a97fc-33c4-495c-8128-33c8d120 Here is one of my result from STAR alignment. As you could see, Uniquely mapped reads: 32.37% and % …
-
How to deal with high % of reads mapped to multiple loci on STAR?
written by Simon Ahn •![enter image description here][1] [1]: /media/images/112a97fc-33c4-495c-8128-33c8d120 Here is one of my result from STAR alignment. As you could see, Uniquely mapped reads: 32.37% and % …
-
Paired-end reads somehow counted twice?
written by Simon Ahn •Hi. I'm new in Bioinformatics and try to extract read counts from fastq files. I compared my result with answer count matrix, and read counts …
-
Tutorial: Slurm and parallelization
written by Juke34Using parallelization in a proper way is not always straightforward. Here some help to better understand how it works with Slurm: (Full part of this …
you asked many questions that got answers. Please review all your previous questions and validate the good answers . e.g: Questions about bam to fastq ; Differential Expression gene In Rna-Seq Experiment using mt.maxT ; get identical number of read1 and read2 aligned ; Parse allele database ; Have you ever obtained user guide of ExomeCNV? ; etc...
I reviewed and answered their comments. What do you mean?
What Pierre means is that you may want to consider upvoting and/or accepting the posts if they have been helpful to you. This gives an incentive to volunteers to share their expertise - although most of us don't do it for the votes - and also provides closure to threads. When others read these threads a year from now it is easier to find helpful comments if they are upvoted or accepted.
OMG. I got the function now. Many thanks!
To me this seems like a problem outside of your CPUs. If there is significant disk reading, and especially if disk writing, that your program is doing, that will be a bottleneck. In such a case it doesn't matter that you have plenty of threads, as they will be running with less than 100% because the disk is not fast enough to provide enough data (or offload the data).
I suggest you try a smaller number, maybe 4-5 threads, and see if that will get them all to run at 100%.
By the way, running "regular" programs as a
rootuser is not good practice.Many thanks,
I'll decrease the threads to run tools, and check the Disk I/O problem in my ubuntu server.
Best,
Oh.
glanceswhich can help you to work out IO bottlenecks on linux.I'll do that from now on. I'll use
glancesand check it. Thank you!