profile and optimize ApplyBQSR

[akiezun]

the goal is to be at least same as gatk3.4 on single thread. This is for the walker version of the tool.
The ticket can be split into a) profile and b) optimize if needed

[akiezun]

Case1:
chr10 of CEUTrio.HiSeq.WEx.b37.NA12892.bam from the gatk bundle, size 1.2Gb

running on Mac OS X 10.9.5 x86_64; Java HotSpot(TM) 64-Bit Server VM 1.8.0_25-b17, local SSD

11.30 min GATK v3.4-46-gbc02625
9 min GATK4 .pre-alpha-51-ge098c21

time for GATK3

real    11m21.538s
user    17m24.320s
sys 0m8.073s

gatk4

real    9m7.691s
user    9m2.250s
sys 0m22.603s

Profile shows that almost all time goes into the bqsr work and nothing obvious can be removed

[droazen]

Can you repeat this same test over an NFS filesystem from a Broad server?

[akiezun]

case 2, on NFS
ref /humgen/1kg/reference/human_g1k_v37.fasta
bam /xchip/cga_home/akiezun/data/CEUTrio.HiSeq.WEx.b37.NA12892.chr10.bam (1.9Gb)

Running on host dataflow01
Linux 2.6.32-573.3.1.el6.x86_64 amd64; Java HotSpot(TM) 64-Bit Server VM 1.8.0-b132.

12.47 minutes GATK v3.4-46-gbc02625
11.8 minutes GATK 4.pre-alpha-45-g168cd60

gatk3 time

real    12m32.499s
user    18m58.643s

gatk4 time

real    11m49.027s
user    12m20.699s