recently I have been dealing with a batch of silkworm(Bombyx mori) RNAseq data. An error arose which I cannot debug. Below is my workflow.
1.the genomic sequence of the silkworm (silkDB 3.0) is about 468.3Mb,
2.The Linux server I am using has 288 cores and 1Tb memory size.
3.No problem arose when I created INDEX files with
4.An error always exits when hisat2 alignment. The following is an example. The memory size usage (%mem) continued to increase after the job submitted.
hisat2 -t -p 30 --dta -x /home/RNAseq_2/source/silkworm/index/silkworm_tran -1 /data/storage04/RNAseq_2/silkworm/majorbio/data4antivirus/cleandata/306D3D1a_R1-clean.fastq.gz -2 /data/storage04/RNAseq_2/silkworm/majorbio/data4antivirus/cleandata/306D3D1a_R2-clean.fastq.gz -S /data/storage04/RNAseq_2/silkworm/majorbio/data4antivirus/alignedFromHisat2Results/306D3D1a.sam
5.the size of the targeted sam file is expected to be 22Gb. But now, %mem is 55 when the sam file is just 7.8Gb.
6.I had tried to run similar 5 jobs with 8cores/job，resulting in the following error message:
(ERR): hisat2-align died with signal 9 (KILL)
I have googled a lot without any progress. Could you please figure out the issue and speed up the job?
Thanks in advance,