-
Notifications
You must be signed in to change notification settings - Fork 103
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Samtools + libdeflate
out performs sambamba
on a single thread
#485
Labels
Comments
Samtools + libdeflate
out performs sambamba
on a single thread
Thanks. It is worth trying and should not be hard to test with guix. |
I heard back on my post to
Anyway, just wanted to report what I found in case it could be useful. |
Oh, meant to include a link to the post at |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hello,
I recently heard about
sambamba
and it's performance gains oversamtools
, and was excited to compare it tosamtools + zlib
andsamtools + libdeflate
(I had also heard thatlibdeflate
really improvessamtools
performance).I compared all three configurations and you can see my full post here: Samtools sort: most efficient memory and thread settings for many samples on a cluster
In short, I compare overall performance (measured by time) at different CPU and memory options. I was impressed that
sambamba
outperforms the other two in pretty much every configuration. There were two things I wanted to share directly that may be of interest:samtools + libdeflate
out performssambamba
which suggestssambamba
could be optimized even more at the compression steps (Fig. 1). You can comparesambamba
(red) andsamtools + libdeflate
(purple) at 1 CPU on the far left of Fig. 1. I'm not sure whatsambamba
uses for compression, though. I'm guessing it doesn't uselibdeflate
, otherwise I suspect it would have suffered from the same poor CPU utilization thatsamtools + libdeflate
suffered from with additional threads. Ifsambamba
is usingzlib
, however, I suspect you could really push the limits for manipulating.bam
files.sambamba
does the best at utilizing allotted CPUs, but it also eventually flattens out. This is obviously a classic computer science problem, but thought you might like to see wheresambamba
flattens out. TBH, I doubt there's much incentive to optimize CPU usage any higher than 9 CPUs, anyway, but who knows?samtools + libdeflate
flattens out very quickly and is unable to fully utilize allotted CPUs as well as the other two configurations (Fig. 2). I assume this boils down tolibdeflate
, but maybe it's more complicated than that. I reported this on thelibdeflate
GitHub page so they can look into it.And thank you for your work. We need more efficient tools like
sambamba
!Figure 1: Realtime vs CPU and Mem Per Thread for
![Realtime vs CPU and Mem Per Thread](https://user-images.githubusercontent.com/1953713/155853889-8768707d-d4bd-47d4-8c24-2b26ec1fd2a4.png)
samtools + zlib
,samtools + libdeflate
(Lsamtools), andsambamba
Figure 2: Requested CPUs vs. CPU utilization for
![CPUs vs. CPU utilization](https://user-images.githubusercontent.com/1953713/155853689-943d6fb0-c7e6-4371-bc70-57df27329624.png)
samtools + zlib
,samtools + libdeflate
(Lsamtools), andsambamba
The text was updated successfully, but these errors were encountered: