VM write performance slow

jhonnytay

Member
Aug 24, 2020
22
0
21
41
Hi Sir,

Would like to know why VM write performance are very slow:

dd if=/dev/zero of=/tmp/test2.img bs=512 count=1000 oflag=dsync
1000+0 records in
1000+0 records out
512000 bytes (512 kB, 500 KiB) copied, 2.22336 s, 230 kB/s



The VM is provisioned under iscsi-lvm storage. With 10G NIC for end to end connection.

Best Regards,
Jhonny
 
dd if=/dev/zero is really not a good way to benchmark. (because it's sequential, and it's writing zero).
alors bs=512 seem quite low to simulate real workload. (maybe try with 4k block for iops bench, and bigger block like 1M for throughput bench)

what is your physical storage ? (local? network ? hdd ? sdd ? how many disk ?)

can you try to bench with fio
(add a second disk for benchmark)


Code:
sequential write with big 4M block (throughput)
fio --filename=/dev/sdb --ioengine=libaio --direct=1 --rw=write --blocksize=4M --iodepth=128 --name=benchmark 


random write with small 4k block (iops)
fio --filename=/dev/sdb --ioengine=libaio --direct=1 --rw=randwrite --blocksize=4k --iodepth=128 --name=benchmark 


random read with small 4k block (iops)
fio --filename=/dev/sdb --ioengine=libaio --direct=1 --rw=randread --blocksize=4k --iodepth=128 --name=benchmark 


sequential write with bug 4M block (throuhput)

fio --filename=/dev/sdb --ioengine=libaio --direct=1 --rw=read --blocksize=4M --iodepth=128 --name=benchmark
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!