Can anyone help troubleshoot my network performance?

[df171422]

I have a 2.4 GB/s WiFi connection on one end, and an Asustor NAS AS6404T with both gigabit ports link aggregated with adaptive load balancing on the other end, and gigabit ports with CAT-6 cables everywhere in between. I'm only seeing 11.4 MB/s transfer speed on average. Can anyone help me figure out why I'm not getting much greater network throughput?

NAS: Asustor NAS AS6404T
ADM: 4.1.0.RLQ1
Both gigabit ports link aggregated with adaptive load balancing

Actual speed:

Actual speed.jpg (31.24 KiB) Viewed 2522 times

Network topology:

Network topology.jpg (29.11 KiB) Viewed 2522 times

WiFi connection:

WiFi connection speed.jpg (42.86 KiB) Viewed 2522 times

ASUSTOR NICs:

ASUSTOR adaptive load balanced NIC.jpg (36.43 KiB) Viewed 2522 times

Many thanks!
Brad

[orion]

Why not try the wired connection on your laptop?

[df171422]

Hello-

Please see below...disabling the WiFi connection and hard-wiring the laptop to the hub didn't make any difference.

WiFi disabled.jpg (14.33 KiB) Viewed 2473 times

Wired actual speed.jpg (30.34 KiB) Viewed 2473 times

Thanks-
Brad

[orion]

Then, you should check link speed on each port (laptop, hub, NAS).
And I think you should disable link aggregation first. Apply one port to check bottleneck first. After all, the throughput is much lower than one giga bit port.

[df171422]

Thanks for the reply. How do I check the link speed on each port? I thought that's what I was doing when I was looking at my WiFi NIC status and the file transfer speed. I'm not sure how to check it from ADM. I'm a software engineer, and I'm dangling on the edge of my networking knowledge.

Thanks again-
Brad

[orion]

You can check under ADM web -> Settings -> Network -> Network interface. It should show you the link speed.

I thought you should be an expert for network stuffs (because you choose to setup link aggregation)... Anyway I don't recommend you to setup link aggregation if you are not familiar with it.

On the other hand, what's your disk configuration, RAID-5? If your disks are very old (bad throughput), it could be a bottleneck. However 11MB/s is quite low even if disks are antiques. Anyway, it's a check point.

[father.mande]

Hi,

If I look your schema ... you use a TP-Link not manageable, nor supporting agrégation (of any type) and not managing STP (spawning tree) protocol

So having 2 links from same machine to the switch, if you are not in separate sub network or in network namespace ... can generate loop and broadcast storm (normally can be reduce by the STP to limit the access to one link.

my suggestion is to try first with only one link to the switch.

If you have network expertise ... you can use tools from Entware APKG to get more information on traffic, capture it and analyze (iptraf, nload, vnstat2, ifstat, etc. etc.) some tools are present in A.D.M. but sometime with less functions.
You can also get some tools like lftp able to test using multi-thread FTP the bandwidth really used ... but SMB is enough in lot of cases.

Philippe.

[df171422]

Thanks for the responses. To orion:

Here's the only info in ADM re: the network interface:

Asustor network interface.jpg (38.03 KiB) Viewed 2313 times

I have 4 Seagate IronWolf 12TB ST12000VN0007 7200 RPM 256 MB Cache drives configured in RAID-10:

Asustor RAID level.jpg (7.63 KiB) Viewed 2313 times

To Philippe:

I have link aggregation configured in Adaptive Load Balancing mode:

Asustor link aggregation mode.jpg (21.81 KiB) Viewed 2313 times

And according to the Asustor documentation, it shouldn't need a managed switch:

Adaptive Load Balancing: Includes transmit load balancing plus receive load balancing for IPV4 traffic and does not require any special switch support. The receive load balancing is achieved by ARP negotiation. This mode provides fault tolerance and load balancing.

https://www.asustor.com/en/online/Colle ... ?topic=307

Something changed - I have no idea what, but I'm seeing a little better throughput with and without link aggregation now. It doesn't matter which I test - both ways are giving me the same throughput:

Actual speed without link aggregation.jpg (29.26 KiB) Viewed 2313 times

Thanks for the help - I really appreciate it.

-Brad