A set of tests run against an idle StoreVirtual 3200 multisite solution with 26x10K SAS drives and a dedicated 1Gbit link be- tween the sites. Round-trip time between the sites is measured to be below 1ms. Files are deleted between each run.
# cat /etc/debian_version
8.8
# lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 4
On-line CPU(s) list: 0-3
Thread(s) per core: 1
Core(s) per socket: 2
Socket(s): 2
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 45
Model name: Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz
Stepping: 2
CPU MHz: 2397.223
BogoMIPS: 4794.44
Hypervisor vendor: VMware
Virtualization type: full
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 15360K
NUMA node0 CPU(s): 0-3
# free -m
total used free shared buffers cached
Mem: 2010 386 1623 5 52 235
-/+ buffers/cache: 99 1910
Swap: 713 0 713
# lsblk -f /dev/sdb1
NAME FSTYPE LABEL UUID MOUNTPOINT
sdb1 ext4 ef9c562a-8db4-4923-918a-1e85a00fc683 /mnt/nch-tests
# fio-v
fio-2.1.11
# gnuplot -V
gnuplot 4.6 patchlevel 6
fio --output=log/4k/4k.out config/4k.fio
fio --output=log/32k/32k.out config/32k.fio
fio --output=log/64k/64k.out config/64k.fio
fio --output=log/128k/128k.out config/128k.fio
for IOPS_LOG in $(echo *_iops*.log); do fio2gnuplot -g -i -p ${IOPS_LOG} -d ../../plot/4k; done
for IOPS_LOG in $(echo *_iops*.log); do fio2gnuplot -g -i -p ${IOPS_LOG} -d ../../plot/32k; done
for IOPS_LOG in $(echo *_iops*.log); do fio2gnuplot -g -i -p ${IOPS_LOG} -d ../../plot/64k; done
for IOPS_LOG in $(echo *_iops*.log); do fio2gnuplot -g -i -p ${IOPS_LOG} -d ../../plot/128k; done