Link Aggregation in Action

In order to get an idea of how link aggregation really helps, we first set up the NAS with just a single active network link. The first set of tests downloads the Blu-ray folder from the NAS starting with the PC connected to port 3, followed by simultaneous download of two different copies of the content from the NAS to the PCs connected to ports 3 and 4. The same concept is extended to three simultaneous downloads via ports 3, 4 and 5. A similar set of tests is run to evaluate the uplink part (i.e, data moves from the PCs to the NAS). The final set of tests involve simultaneous upload and download activities from the different PCs in the setup.

The upload and download speeds of the wired NICs on the PCs were monitored and graphed. This gives an idea of the maximum possible throughput from the NAS's viewpoint and also enables us to check if link aggregation works as intended.

The above graph shows that the download and upload links are limited to under 1 Gbps (taking into account the transfer inefficiencies introduced by various small files in the folder). However, the full duplex 1 Gbps nature of the NAS network link enables greater than 1 Gbps throughput when handling simultaneous uploads and downloads.

In our second wired experiment, we teamed the ports on the NAS with the default options (other than explicitly changing the teaming type to 802.3ad LACP). This left the hash type at Layer 2. Running our transfer experiments showed that there was no improvement over the single link results from the previous test.

In our test setup / configuration, Layer 2 as the transmit hash policy turned out to be ineffective. Readers interested in understanding more about the transmit hash policies which determine the distribution of traffic across the different physical ports in a team should peruse the Linux kernel documentation here (search for the description of the parameter 'xmit_hash_policy').

After altering the hash policy to Layer 2 + 3 in the ReadyNAS UI, the effectiveness of link aggregation became evident.

In the pure download case with two PCs, we can see each of the PCs getting around 800 Mbps (implying that the NAS was sending out data on both the physical NICs in the team). An interesting aspect to note in the pure download case with three PCs is that Machine 1 (connected to port 3) manages the same 800 Mbps as in the previous case, but the download rates on Machines 2 and 3 (connected to ports 4 and 5) add up to a similar amount. This shows that the the network ports 4 and 5 are bottlenecked by a 1 Gbps connection to the switch chip handling the link aggregation ports. This is also the reason why Netgear suggests using port 3 as one of the ports for handling the data transfer to/from the link aggregated ports. Simultaneous uploads and downloads also show some improvement, but the pure upload case is not any different from the single link case. These could be attributed to the limitations of the NAS itself. Note that we are using real-world data streams transferred using the OS file copy tools (and not artificial benchmarking programs) in these experiments.

Introduction and Benchmarking Setup The Promise of Gigabit Wi-Fi and Concluding Remarks
POST A COMMENT

67 Comments

View All Comments

  • Ertaz - Thursday, December 31, 2015 - link

    Hey, thank you for taking the time to do this review. This is good info and it's given me things to think about when I upgrade my network in the middle of the year. Reply
  • creed3020 - Thursday, December 31, 2015 - link

    Thanks for confirming my suspicions, at least with this hardware, that wireless AC at the highest end is still not ready to replace my Cat5e cables snaked through the house.

    I love my fast connection to my NAS, router, and other clients without high latency and jittery speeds.
    Reply
  • IndianaKrom - Thursday, December 31, 2015 - link

    The other thing to keep in mind is that all 802.11 based standards are half-duplex and all these routers claims of bandwidth is the sum of the total system bandwidth in both directions. But in reality the bandwidth is split 50/50 between the transmit and receive sides of the time division multiplexing. So in any one direction transfer the maximum theoretical throughput of a pair of wireless devices is 50% of the negotiated link speed.

    Basically if wired lan was advertised the same as wifi routers, then your 1 gigabit cat5e would be marketed as "2 gigabits".
    Reply
  • zodiacfml - Thursday, December 31, 2015 - link

    Wow, I was about the mention the same thing. I did a lot of reading last year on Wi-Fi and standards that I came to know of this half-duplex operation as I am perplexed why multi-gigabit Wi-Fi don't come close LAN cabling. Even uglier are the spatial streams which has diminishing returns. A second spatial stream, if I'm correct, is around 50% improvement and the third stream is around 20 to 30%. Reply
  • easp - Friday, January 1, 2016 - link

    You make a valid point about the difference between full and half-duplex operation.

    You miss though that ethernet was originally shared and half-duplex, and so a 10baseT or 10base2 network segment shared 10mbps among all connected devices. This bandwidth was measured at the PHY (physical) layer, and didn't take into account the overhead imposed the collision detection and avoidance MAC (media access control). During the the rise in full-duplex NICs and switches starting in the mid-1990s, bi-directional communication was still a selling point.

    When WiFi arrived, it was specced similarly to early ethernet. The quoted speeds were based on the maximum available PHY rate, and ignored the overhead imposed by other layers of the wireless stack. This pattern held until 802.11ac, when the quoted datarates were at the MAC layer, and arrived with efficiency improvements in the MAC. However, they still are half-duplex.

    While I agree that the distinction between the full-duplex norms of most wired networking and the half-duplex norms of WiFi are important to consider. I think you've manage to both under and overstate the implications when it comes to WiFi.

    Also, your comment on shared bandwidth between two wireless stations connecting through the same WiFi router/access point should probably acknowledge that things change straightforward once multi-user MIMO (MU-MIMO) becomes better supported and adopted, particularly since the article you are commenting on is about MU-MIMO capable hardware.
    Reply
  • easp - Friday, January 1, 2016 - link

    Derp. I meant to cut my second-to-last paragraph. Reply
  • Notmyusualid - Sunday, January 3, 2016 - link

    Add to your half-duplex wireless hates, air listening times, back-off times, re-tranmissions, noisy environments (I had 57 viewable networks from the apartment in Copenhagen), and more all lead to increased latency. So I gave up on 802.11g long ago... 2.4GHz was saturated.

    My two C3750G switches are hard-wiring my house just fine, without LACP or PAGP, my etherchannels are hard coded as ON.

    But I actually still like this product - I've seen more than enough small business run on non-enterprise gear (NOT my decision), and so having etherchannels available to a NAS makes sense - for a storage solution that might be under quite some load with a dozen or so users.

    And wait until DD-WRT gets their hands on it, you might get some port security etc too!
    Reply
  • michaelag - Thursday, December 31, 2015 - link

    "On the other hand, 802.11c enables router manufacturers to market multi-gigabit Wi-Fi."
    Correction, "802.11ac"?
    Reply
  • ganeshts - Thursday, December 31, 2015 - link

    Thanks for pointing out. Fixed the typo. Reply
  • The_Assimilator - Thursday, December 31, 2015 - link

    I'm honestly surprised there hasn't been a lawsuit in the US regarding the lies about wireless speeds that router companies peddle. "5.3Gbps WiFi Speeds"... I doubt Netgear has ever even seen that in their own testing, and I doubt anyone ever will.

    Maybe one day wireless speeds won't be "up to" with the average maximum throughput being 20% of what the manufacturer claims... and maybe I'll fart butterflies. Until then, I'm sticking with wired Ethernet - it may be an old standard, and a pain, but at least it delivers.
    Reply

Log in

Don't have an account? Sign up now