I am trying to run PBDAGCON on the blasr alignment of illumina contigs to pacbio reads through DBG2OLC.
I was testing the tool with one specific alignment (the 1st one). I have been running with strange problems with it.
First, it was running really slow with one thread so I increased threads to 10. Since it had no improvement I played around with buffer size options starting at 256000 to 32, and once it suddenly ran on one alignment in less than 15 sec at rbuf and wbuf of 4096.
I tried to re-run it with the same option, but failed repeatedly on trails with buffers and threads. I do not understand why it ran so fast at one moment, but later I cannot reproduce the result in anyway.
It has been running from more than an hour with the same option now, but still no output.
I have been using a shared server with ~1.1TB memory and 48 cores. Is there some reasoning behind this?
