White Papers

6
Test Configurations and Data
Since the main interest of this study is to make a reasonable observation for memory consumption behavior between two systems, all
DIMM slots in both systems were fully populated to match the amount of memory to 1.5 TB. Although other components can influence
the benchmark results like storage configuration, those influences should be marginal since the performance of de novo assembly
applications is largely bounded by CPU/memory. The details of test configurations are listed in Table 1.
Table 1Test configurations
PowerEdge R940
PowerEdge R930
CPUs
4x Intel
®
Xeon
®
Platinum 8180M
4x Intel
®
Xeon
®
CPU E7-8890 v3
Base Frequency (GHz)
2.5
# of Cores
28, total of 112
16, total of 64
TDP (W)
205
165
Memory
48x 32GB DDR4-2666MHz, total of 1.5 TB
96x 16GB DDR4-2400MHz, total of 1.5 TB
Storage
12x 1.9TB 12 Gps SSDs in RAID 0
Boot disk is 300GB RAID 1 with 1 spare (total of
3 disks), 3 TB RAID 0 based on 2x Intel
SSDSC2BX01 SSD 1.5TB
Kernel
3.10.0-514.el7.x86_64
Operating System
Red Hat Enterprise Linux Server release 7.3 (Maipo)
The choice of test data is a whole human genome sequencing data set, ERR318658 containing 3.2 billion reads (10). The number of
reads in this whole genome sequencing data belongs to the range of deep sequencing. For the benchmark purpose, smaller data sets
were generated from ERR318658 containing 1.6 billion reads, 160 million reads, and 80 million reads by truncate the original
sequencing files.
Performance with SOAPdenovo2
SOAPdenovo2 is specially designed for human sized genomes and assembles Illumina Genome Analyzer short reads (11). The
program creates new opportunities for building reference sequences and carrying out accurate analysis of unexplored genomes in a
cost effective way.
Running Time
Figure 3 shows wall clock running times with various sizes of input data with 62 cores. In order to make a fair comparisons, two cores in
R930 was unused to avoid unexpected performance degradation. PowerEdge R940 can process 3.2 billion read data 27% less time
than PowerEdge R930 in core to core comparison.
Figure 3 Running time comparisons with 62 cores: the total number
of cores available in R930 is 64 while 112 cores are available in
R940. The choice of 62 cores is to obtain a fair comparison
between two systems. Two cores are kept for the system in R930 to
avoid any unexpected performance degradation