Abacus stops loading prot.xml files

[Alexander Rabe]

Dear all,

I would like to know the following: Is there a limited number of files Abacus can deal with?

The problem is the following:

My set of data consists of 324 samples (162 prot.xml, 162 pep.xml files with 150 - 200 MB each). Abacus starts to load the prot.xml files, but when it reaches the 129th prot.xml file it stopps working. I run Abacus on a virtual machine with 12 CPU's and 50 GB of Memory.

If there is no limitation, could it be that I still need more Memory?

Thank you very much for your help.

Kind regards,


Alexander

[Damian]

Hi. 

There is no memory limit on Abacus. Although the HSQLDB jar (which is used in Abacus) may have one. If it always stops at the same file that is informative. 

Could you post the screen output up until the point where the program crashes?

Damian

--
You received this message because you are subscribed to the Google Groups "Abacus Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to abacus-suppor...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Alexander Rabe]

Hi Damian,

I tried to analyze the data several times with Abacus over the weekend, but it was always the same as described before. I have made four pictures. Hopefully, they are helpful.
On picture "Abacus_no_more_parsing.jpg" you can see, that the CPU Usage is about 7 %, but before it was that low the CPU Usage was between 80 % and 99 % for a while. After 2 or 3 minutes the CPU Usage dropped down to 7 %.

Alexander

[GATTACA]

Could you run Abacus from the command line instead of double clicking on the JAR file?

The error messages are printed to the command line but you won't see them unless you call the jar from there.

Something like: java -Xmx48g -jar abacus.jar should work from the directory  within the directory where the jar file is found.

You then run the program like you normally would and when it crashes the errors will appear in the command line window. This is the information I need to begin to decipher you problem.

Thanks,

Damian

[Alexander Rabe]

Hi Damian,

I am sorry that my answer took so long. I have started Abacus from the command line and it worked. Because of the huge data set I had to increase the number of CPU's and the size of memory and there were no problem at all anymore. I have no idea why it worked better using the command line (maybe you have an explanation), but now every thing is fine and I can start analyzing my data.
Thanks a lot for your help so far.

Kind regards,

Alexander

[Damian]

Glad to hear it. 

Just wanted to let you know that Abacus isn't multithread optimized so throwing more CPUs at it won't make huge a difference. Extra memory should help a lot. 

As for why you got it to work on the command line that maybe coincidence. I suspect increasing the memory is what made the difference. 

Cheers

Damian

Sent from my PDA. 

Please excuse typos & brevity. 

--