I know this has been discussed multiple times but I have not found any solution so far that worked so posting here hoping there is some solutions in Dec 2021…
I have a Dell R640 server with dual Xeon Gold processors and 384gb ram. The chassis is only sata/sas drive (does not support u.2) and don’t have budget for new server that supports u.2.
Note - my use case is to provide storage for VM to take advantage of NVMe speeds.
So we opted for PCI card - Dell SSD NVMe M.2 PCI-e 2x Solid State Storage Adapter Card 23PX6 NTRCY. It supports 2 NVME drives and connects via bifurcation to both as x4 PCI lane.
I have two Kingston 2TB nvme drives and I created mdadm based RAID1.
The write performance of a single nvme ssd is 1800MBps. But the RAID1 has write speed of 500MBps.
I found that the Bitmap= Internal was possible problem, and I applied
mdadm <dev> --grow --bitmap=none
Even after this the performance is nearly the same.
Any suggestions on what else I can try?
So I am not sure what happened - today when I ran the speed test again, the speed is within expectations -- Read of 1039 MBps and write of 1352MBps (using crystaldiskmark on a VM on this host)
mdadm --detail /dev/md0
/dev/md0:
Version : 1.2
Creation Time : Sun Nov 28 19:08:22 2021
Raid Level : raid1
Array Size : 1953381440 (1862.89 GiB 2000.26 GB)
Used Dev Size : 1953381440 (1862.89 GiB 2000.26 GB)
Raid Devices : 2
Total Devices : 2
Persistence : Superblock is persistent
Intent Bitmap : Internal
Update Time : Thu Dec 2 10:33:50 2021
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Consistency Policy : bitmap
Name : server1:0 (local to host server1)
UUID : 69bab65f:9daa6546:687fc567:bd50164a
Events : 26478
Number Major Minor RaidDevice State
0 259 2 0 active sync /dev/nvme0n1p1
1 259 3 1 active sync /dev/nvme1n1p1