DS1817+ Poort Performance

IF your questions pertains to using the file server performance, transfer issues, jumbo frames tuning, post it here.
Forum rules
1) This is a user forum for Synology users to share experience/help out each other: if you need direct assistance from the Synology technical support team, please use the following form and check this FAQ:
https://myds.synology.com/support/suppo ... p?lang=enu
https://www.synology.com/en-global/know ... se/faq/586
2) To avoid putting users' DiskStation at risk, please don't paste links to any patches provided by our Support team as we will systematically remove them. Our Support team will provide the correct patch for your DiskStation model.
Randvegeta
I'm New!
I'm New!
Posts: 2
Joined: Tue Nov 21, 2017 4:28 pm

DS1817+ Poort Performance

Postby Randvegeta » Tue Nov 21, 2017 4:44 pm

Hello all.

Couple of weeks ago I purchased 2x DS1817+ with the optional dual 10G SPF+ cards.

I'm looking to deploy these as a high availability storage cluster for a small Xen HA virtualization cluster.

One of the DS1817+ units (Call it NAS A)is configured as so:

6x4TB HDD (RAID 10, ext4)
2x480GB SSD (RAID 1 SSD Caching for 10TB RAID 10 volume)

The second DS1817+ (call it NAS B)has 6x3TB HDD in RAID 6, which has the same amount of usable disk space as the RAID 10 array. No SSD caching.

NAS A and B are directly connected to each other via the 2nd 10G port. They both connect to a 10G switch of which the Xen nodes are also connected.

The idea is to use NAS A as the 'primary' storage and NAS B as the backup, in case NAS A fails in some way. NAS A and B are supposed to be configured in SHA mode.

I have yet to actually put the Synology NASs into high availability clustering yet as I am still testing the performance and I am getting some very strange, and generally quite poor results.

Basic tests directly on NAS A (connected through SSH) shows I can write to the RAID 10 array at about 200-400MB/s. The performance is very uneven and the performance does not improve with SSD caching activated. I have created an ISCSI LUN and connected it to a Xen node where I created a VM where the virtual disk is hosted on on the NAS. Performance varied from 50-110MB/s. Which is worse than local disk, and less than 1/4 the speed I am seeing directly on the NAS itself.

I decided to test NAS B, which is in RAID 6. Other than the RAID type, they are configured in the same way. So I created the same kind of iSCSI LUN, same size and same settings. Then created a VM on the same node. This time, the write speed of the VM to the virtual disk is about 350MB/s. 3-7x faster than the RAID 10 version.

Why on earth would the RAID 6 perform better than the RAID 10 array on the same NAS model? Connected to the same switch, running VMs on the same node.

I was expecting speeds of about 400-600MB/s on the RAID 10 array, and considerably less performance on the RAID 6.

Any idea what's going on?

Return to “File Server Performance”

Who is online

Users browsing this forum: No registered users and 4 guests