Am I doing something terribly wrong, or is SpeedFusion bonding useless?

Product Discussion Pepwave MAX
4

Hi Shane, please open a support ticket so we can investigate further:
https://contact.peplink.com/secure/create-support-ticket.html

5

Ticket 758708 has been open on this for almost a week. A recent speedtest showed 1.2 Mbit down, 4.1 Mbit up with SpeedFusion. Unchecking “send all traffic” to that VPN gets me 35 Mbit down, 13 Mbit Up! I must be doing something very, very wrong.

6

Hi,

We are working on the ticket. As we need to perform full performance test to confirm the issue

Thank You

7

A follow-up for anyone who may be reading this: it seems a resolution was achieved (see below). We are using this in the field in 1 week for a real-world production, so we’ll be able to tell then…but the immediate results are highly encouraging. Apparently a mismatch on MTU between HD4 and FusionHub. Interestingly, I did not change MTU on either device (don’t even see the setting on HD4 web interface, but perhaps it is there). SpeedFusion defaults to 1440, but they changed it to 1428 to solve. Perhaps something to look at if you have SpeedFusion issues with HD4/FusionHub. See ticket resolution from Peplink below:

We have identified the issue and fixed the setting. The SpeedFusion slow performance are due to the MTU miss match between Max HD4 & the FusionHub. I have changed the MTU size for the FusionHub and this should fixed the performance issue.

Initial setting: MTU 1440
Changed: MTU 1428 (Refer to the attached screenshot MTU.PNG)

Can you please redo the speedtest again ?

Base on the PepVPN test tools you should get the below:
MAX HD4 SF Upload speed: 20-27Mbps
MAX HD4 SF Download speed: ~ 50Mbps

1 Like
8

Hi,

The SpeedFusion slow performance are due to the MTU miss match between Max HD4 & the FusionHub. This is a known issue for the FusionHub version 6.2.2 and targeted will be fix in FusionHub version 6.3.0.

Workaround for this is to define the FusionHub WAN interface MTU size to follow the lowest MTU size that defined for the MAX HD4 cellular WANs

For more information regarding to the FusionHub WAN interface MTU setting, please refer to the attached screenshot.


Thank You

9

What kind of bonding efficiency where you able to achieve with your HD4? You said you had 4 modems plugged in, did you ever see how much faster the speedfusion made your bond?

10

Hi TD,

For the support case:

Without SF Bonding:13 Mbps Upload, 35 Download
MAX HD4 SF Bonding: 20-27Mbps Upload, ~ 50Mbps Download (This is limited by FusionHub License).

Cellular WAN speeds/Characteristic are variants from time to time, so you have to take consideration for the available bandwidth before you compare again with the SF bonding performance.

For more information, please refer to the URL below:

Thank You

11

Just an update to this. We rarely use HD4 at all because it never seems to do much for our throughput. Fired it up today, haven’t changed any settings…it just performs as it does in my first posts. I get faster upload/download WITHOUT going through SpeedFusion than I do with. Pretty frustrating.

12

Hi Shane,

Just wanted to confirm the setup you mentioned here is same as what you did previously (adjusting the mismatch MTU between HD4 and FusionHub) as your post early this year?

Thanks and regards.

13

MTU is the same on FusionHub and HD4 cell WANs (I believe 1428). Havne’t changed a thing.

14

Hi Shane,

This is the same device which reported in RT#758708? If so, please enable Remote Assistance and we will follow from there.

Thank you.

15

Hi guys,

How did you finally solve the problem? Any follow up?

Thanks a lot,
Jarek

16

acompanhando…

17

I would suggest to go through the following URLs to verify the SF performance issue.

18

Hi - I mistakenly added this along with another reply to a different post. That was not the topic I intended to post to, please accept my apologies for unintentionally “spammin” the forum.

Does this sound similar to what anyone else is seeing?
Looking at https://download.peplink.com/partner/SpeedFusion_Best_Practices.pdf
Page 4 describes the use of IMIX (RFC 6985 - IMIX Genome: Specification of Variable Packet Sizes for Additional Testing) traffic flows.
Page 5 describes using IMIX over a multiple WAN speedfusion, with the example Download: 10Mb + 10Mb = 20Mbps – 19% = 16.2Mbps
That sounds excellent.

Our setup with SpeedFusion between two Balance 710s (fw ver 7.1.0) has one B710 with a 500Mbps Internet connection, and the other B710 has 5 WAN Internet circuits:
All WAN Internet circuits have been tested to rated capacity individually.
Download: 100Mbps (W1) +60Mbps (W2)+ 60Mbps (W3) + 60Mbps (W4) + 60Mbps (W5), but the throughput using speedtest.net is 35Mbps.

Does that look right?

Thanks
Dana

1 Like
19

Not immediately no.

  • What speeds do you see over the SpeedFUsion VPN for TCP & UDP tests in the PepVPN test tool?
  • When you run a speedtest what happens to individual WAN latencies?
  • What technology are the WAN links?
1 Like
21

Hi,
Have you found a solution to your speedfusion problems ? I’m in the same case as you…
Thanks for your feedback
Erik

22

@mssfrance, this has been answered below.

In short, you need to understand more on the WAN links you used and do the appropriate throughput testing. You may get help from your Peplink partner if needed.

1 Like
23

Dear All,

I have the same issue here:

  • two LTE connections
  • different performance
  • the result is only the avarge speed of both not the total of both ( minus overhead)
  • what is strange that upload seems to be limited to 50 MBps

See screenshot. I hope it is self-explaining:

image
image955×674 36.9 KB

Strange that there are so many abadonned threads on this. I hope someone has a solution idea.

24

There is a lot going on in that graph.
Lets work our way through it.

image
image953×675 60.9 KB

Download test #1. Cell1 and Cell2
1.~25MBps download. Cell1 doing most of the work. Cell 2 throughput is low maybe ~5mbps.
2. When combined like this latency is managed and looks reasonable <~100ms
3. Cellular 2 is lossy touching on 50 packets/sec and Cellular1 has a noticeable amount also maybe 25pkt/s

Download Test #2 - Just Cellular 1 (CAT12)
4. Loads of bandwidth ~90+Mbps
5. Latency stays low <100ms
6. Some packet loss but its minimal

Download test #3 just cellular 2 (CAT4)
7. 8-10Mbps?
8. Monster amounts of latency! Over a second! Worse than Satellite.
9. Tiny amount of packetloss at the beginning of the test - then nothing.

Upload Test #1 Combined
10. 30Mbps ish
11. Latency is fairly evenly poor - both connections must be creeping up to 200 - 250ms+
12. So much packet loss per second on both links!

Upload Test #2 Just Cellular 1 (CAT12)
13. Touching 45Mbps of bandwidth which is good
14. Latency isn’t great in the region of 200ms
15. Slight touch of packetloss

Upload Test #3 - Just Cellular2 (CAT4)
16. 30Mbps+ bandwidth
17. Latency is not good >250ms at times.
18. Chunk packet loss.

Some observations / recommendations.

  • We can see from download test #1 and upload test #1 that the two cellular connections are interfering with each other. When both cellular connections are in use, we see high packetloss on both connections (3 & 12).

Check antenna positions, make sure you have all antennas connected and separated well. Also what frequencies are each cellular connection using? Are they on the same bands? Are you using splitters or amplifiers maybe?

  • Download latency on cellular 2 (CAT4) is horrific (8) that needs investigation. Are you right on the cell edge perhaps? What are the RSRQ RSRP and RSSI values for this connection? Sometimes this can be caused by a mis-configured tower.

If you want to continue to work on this please open another new thread (you can copy and paste these two posts into the top). You have duplicate posted your question across a lot of the threads you found which makes things messy.

2 Likes