Unconfigured Ad

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • tirohia
    Member
    • Nov 2011
    • 47

    #1

    Hisat2 multithreading not working.

    As best I can see, this should be working. I've got 16 cores:

    Code:
    $ grep -c ^processor /proc/cpuinfo     
    16
    I then attempt to use hisat2 with multiple threads. I've tried using both --threads -p
    Code:
    hisat2 --threads 15 -x ../genome/genome -1 fq1.fq -2 fq2.fq -S aligned.sam
    Neither works. If I flick to another screen and run top, it looks like:
    Code:
    top - 07:29:52 up  4:17,  2 users,  load average: 0.61, 0.34, 0.44
    Tasks: 164 total,   2 running, 162 sleeping,   0 stopped,   0 zombie
    %Cpu(s):  2.3 us,  4.0 sy,  0.0 ni, 93.7 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
    KiB Mem:  61976884 total, 61398192 used,   578692 free,    35588 buffers
    KiB Swap:        0 total,        0 used,        0 free. 59440624 cached Mem
    
      PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND         
    10666 tirohia   20   0  982432   3984   3700 R 100.0  0.0   3:28.93 hisat2-align-s  
    10667 tirohia   20   0   25740   2948   2440 R   0.3  0.0   0:00.28 top             
        1 root      20   0   28564   4744   3120 S   0.0  0.0   0:01.46 systemd
        2 root      20   0       0      0      0 S   0.0  0.0   0:00.00 kthreadd
    hisat2 isn't getting past 100% - my understanding of top is that it's percentage is based of the usage of a single cpu - so it should be showing 1500%.

    What am I missing?

    Cheers
    Ben.
  • mastal
    Senior Member
    • Mar 2009
    • 666

    #2
    I haven't used hisat2, but with tophat, not all of the stages of the analysis were multi-threaded. So unless you look at top while hisat2 is running one of the multi-threaded stages, you might only see it using 100% CPU.

    Comment

    • tirohia
      Member
      • Nov 2011
      • 47

      #3
      So when I run an alignment with bowtie2, it takes maybe an hour or. When I do it with Hisat2, it'll run for days.

      I've been logging the CPU usage with:
      Code:
      while true; do (echo "%CPU %MEM ARGS $(date)" && ps -e -o pcpu,pmem,args --sort=pcpu | cut -d" " -f1-5 | tail) >> ps.log; sleep 5; done
      When I run hisat2, the cpu usage stays at about 100%. When I run bowtie2, it shoots up to ~1500%, as I would expect. Which suggests to me that the hisat2 multiple threading section isn't kicking off at all.

      Which, I have to say, leaves me contemplating just returning to bowtie2.

      Comment

      • dpryan
        Devon Ryan
        • Jul 2011
        • 3478

        #4
        Please report this as a bug to the developers. I should note that it runs multithreaded for us, so there's some odd interaction between the binary and your system.

        Comment

        • tirohia
          Member
          • Nov 2011
          • 47

          #5
          Is there a formal bug submission process (if so I can't find it), or does one just submit a new issues via github?

          Comment

          • dpryan
            Devon Ryan
            • Jul 2011
            • 3478

            #6
            Github is usually as formal as things get, so yes, post it there.

            Comment

            Latest Articles

            Collapse

            • SEQadmin2
              New Genomics Technologies Take Aim at Long-Standing Limits
              by SEQadmin2


              Researchers using sequencing and genomics tools often have to make trade-offs. They can choose between speed or scale, short reads or long-range information, or targeted panels or a view of the whole transcriptome. New technologies that have been released this year are built to address those tough choices.

              We asked six companies the same four questions to learn about their latest products. The new technologies bring a lot to the table, including rethinking sequencing
              ...
              Yesterday, 10:25 AM
            • SEQadmin2
              How Immunogenomics Decodes Immunity’s Genetic Blueprint
              by SEQadmin2




              The immune system’s power comes from its genetic diversity, allowing myriad threats to be neutralized through first recognizing foreign antigens. That diversity is also what makes the immune system so difficult to study. Recent advances in sequencing technology and computational biology, however, are giving researchers new tools to understand immune responses and immune-related diseases in greater detail.

              This convergence of genetics, immunology, and computation...
              09-01-2026, 05:41 AM

            ad_right_rmr

            Collapse

            News

            Collapse

            Topics Statistics Last Post
            Started by SEQadmin2, 09-25-2026, 09:06 AM
            0 responses
            29 views
            0 reactions
            Last Post SEQadmin2  
            Started by SEQadmin2, 09-23-2026, 11:05 AM
            0 responses
            24 views
            0 reactions
            Last Post SEQadmin2  
            Started by SEQadmin2, 09-18-2026, 11:37 AM
            1 response
            46 views
            0 reactions
            Last Post pekgio
            by pekgio
             
            Started by SEQadmin2, 09-16-2026, 10:23 AM
            1 response
            55 views
            0 reactions
            Last Post pekgio
            by pekgio
             
            Working...