I am not sure... I just rememberd that I read something about ScaleIO and ECS... But not really sure.
Well thats what he thought too...That definitely will be interesting.
At least with a support contract and officially sanctioned hardware (?) you can bug vmware about the horrible speed
Not that it helps in this situation but it can run well, think 10 nodes with 2 groups a node and all flash... works well but that’s at scale, Low count of nodes is ‘ok’ if you compare performance to old spinning disks only systems but compared to some local NVME you will disappointed
Just found that old thread again here https://forums.servethehome.com/ind...vers-build-starwind-vsan-or-vmware-vsan.8652/
Hehe, I am just doing the same right now. Yes, as far as I see u put it on top of the filesystem...
I installed Win2k16 on a RAID1 I made between the 2x Intel DC S4600. So this is my C drive and the Intel NVME is D drive which I use as decoration... to be honest... since the max link speed I will ever achive... is anyway 20GBit limited to the mezzanine cards inside the compute cluster nodes... I think I can resell the brand new 900p and buy more S4600 drives...
Unfortunately ATTO does only have Queue Depth 2 as the lowest value... But I can run any benchmark you want me to... Just tell me what you want me to run ;-)Could you do me a favour and run with QD=1 ?
I am looking for single user max perf and that would be an interesting data point
#!/bin/bash
env="compuverde_2_nvme_2"
echo "4kreadtest_d1_j1 " >$env
for RUN in $(seq 1 5)
do
./fio --filename=test --direct=1 --rw=randrw --refill_buffers --norandommap --randrepeat=0 --ioengine=libaio --bs=4k --rwmixread=100 --iodepth=1 --numjobs=1 --runtime=60 --group_reporting --name=4kreadtest >>${env}
sleep 20
done
echo "4kwritetest_d1_j1 " >>$env
for RUN in $(seq 1 5)
do
./fio --filename=test --direct=1 --rw=randrw --refill_buffers --norandommap --randrepeat=0 --ioengine=libaio --bs=4k --rwmixread=0 --iodepth=1 --numjobs=1 --runtime=60 --group_reporting --name=4kwritetest >>${env}
sleep 20
done
echo "4kreadtest_d16_j16 " >$env
for RUN in $(seq 1 5)
do
./fio --filename=test --direct=1 --rw=randrw --refill_buffers --norandommap --randrepeat=0 --ioengine=libaio --bs=4k --rwmixread=100 --iodepth=16 --numjobs=16 --runtime=60 --group_reporting --name=4kreadtest >>${env}
sleep 20
done
echo "4kwritetest_d16_j16 " >>$env
for RUN in $(seq 1 5)
do
./fio --filename=test --direct=1 --rw=randrw --refill_buffers --norandommap --randrepeat=0 --ioengine=libaio --bs=4k --rwmixread=0 --iodepth=16 --numjobs=16 --runtime=60 --group_reporting --name=4kwritetest >>${env}
sleep 20
done
