Last edited:
I'm scratching my head on this one as since my last actual test we have had DSM updates, macOS updates and probably a switch update.
The NAS is my RS217, so weak in just about every area, running with 2 x 1GbE links (it only has 2 ethernet ports), no LAG, no Bond, large MTU, SMB3. Volume is a single SSD and no errors or issues reported anywhere. I noticed that it seemed to be running slower on transfers so ran a few tests and double-checked network settings.
For read speed I am getting around 230MB/s (slightly up on last year) but only peaking at around 130MB/s and typically around 115MB/s for writes - close to 1GbE speed. It used to be around 200MB/s for writes and no lower than 180MB/s. Data throughput is being shared equally between the 2 x 1GbE links.
As an example using a typical speedtest app:
DSM7 Resources - not exhausting the (anaemic) CPU or RAM:
DSM7 Settings:
My switch port settings (port 17&18 being used for this NAS):
NAS SMB config:
Can anybody clue me in as to why the link only achieves around 1GbE speeds for writes or spot an error anywhere?
️
The NAS is my RS217, so weak in just about every area, running with 2 x 1GbE links (it only has 2 ethernet ports), no LAG, no Bond, large MTU, SMB3. Volume is a single SSD and no errors or issues reported anywhere. I noticed that it seemed to be running slower on transfers so ran a few tests and double-checked network settings.
For read speed I am getting around 230MB/s (slightly up on last year) but only peaking at around 130MB/s and typically around 115MB/s for writes - close to 1GbE speed. It used to be around 200MB/s for writes and no lower than 180MB/s. Data throughput is being shared equally between the 2 x 1GbE links.
As an example using a typical speedtest app:
DSM7 Resources - not exhausting the (anaemic) CPU or RAM:
DSM7 Settings:
My switch port settings (port 17&18 being used for this NAS):
NAS SMB config:
Code:
root@Dragon:~# cat /etc/samba/smb.conf
# Copyright (c) 2000-2019 Synology Inc. All rights reserved.
## Don't Panic, in big friendly letters ##
[global]
server multi channel support = yes
aio read size = 0
aio write size = 0
printcap name=cups
winbind enum groups=yes
include=/var/tmp/nginx/smb.netbios.aliases.conf
encrypt passwords=yes
min protocol=SMB2_10
security=user
local master=no
realm=*
syno sync dctime=no
passdb backend=smbpasswd
printing=cups
max protocol=SMB3
winbind enum users=yes
load printers=yes
workgroup=WORKGROUP
root@Dragon:~#
Can anybody clue me in as to why the link only achieves around 1GbE speeds for writes or spot an error anywhere?
