High latency over Force300 PTP link

Hi all,

we're having a discussion with some other technician because they says that Force300 PTP links have low latency like it was with Force200 ePTP mode (about 1-2 ms).

From what I've tried all over our Force300 PTP links, I've always seen about 8-10 ms using TDD 50/50 or 75/25 with 40 o 80 Mhz channel... sometimes if we've low traffic, we get maybe 7-8 ms, but under heavy traffic we reach also 10-15 ms!

How do other Force300 PTP links work?

I really hope that there'll be something like an ePTP mode soon over Force300!!!

Regards,

Paolo

1 Like

up.....

no one's getting any experience in low latency ptp link with F300 ???

Do you knonw if they will support some sort of ePTP mode in the near future?

Thanks to everyone that will reply... ;)

Hi. Yes, we also need a lower latency mode for the ePMP3000 and Force300 PTP usage. Good news from what I undestand, a new ePTP mode is coming.

So, the situation as I understand it was this: Because the Force300's regular TDD Mode's latency was so much better than the previous generations, Cambium kinda put the ePTP on the back burner because it would require a complete rewrite, and instead focused on other things first. Everyone needs QOS and stability and other things first, and the regular TDD latency was pretty decent.

Anyway - here is one of our Force300/Force300 PTP links in TDD mode.

On this Force3000 PTP link - the average is 3.8ms - with min of 1.7ms and max of 6.8ms

PING 10.0.0.46 (10.0.0.46) 32(60) bytes of data.
40 bytes from 10.0.0.46: icmp_req=1 ttl=64 time=5.19 ms
40 bytes from 10.0.0.46: icmp_req=2 ttl=64 time=3.12 ms
40 bytes from 10.0.0.46: icmp_req=3 ttl=64 time=4.92 ms
40 bytes from 10.0.0.46: icmp_req=4 ttl=64 time=3.37 ms
40 bytes from 10.0.0.46: icmp_req=5 ttl=64 time=2.02 ms
40 bytes from 10.0.0.46: icmp_req=6 ttl=64 time=6.80 ms
40 bytes from 10.0.0.46: icmp_req=7 ttl=64 time=4.26 ms
40 bytes from 10.0.0.46: icmp_req=8 ttl=64 time=4.55 ms
40 bytes from 10.0.0.46: icmp_req=9 ttl=64 time=1.71 ms
40 bytes from 10.0.0.46: icmp_req=10 ttl=64 time=2.31 ms

--- 10.0.0.46 ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9012ms
rtt min/avg/max/mdev = 1.723/3.877/6.803/1.467 ms 


The problem with the above Force300 is that below is a Force200 link in ePTP mode. Latency on the Force200 in ePTP mode averages only 1.5ms plus/minus 0.2 ms, so almost perfectly flat.

Force200 ePTP - ave of 1.5ms, min of 1.3ms, max of 1.7ms
 

PING 10.2.4.76 (10.2.4.76) 32(60) bytes of data.
40 bytes from 10.2.4.76: icmp_seq=1 ttl=64 time=1.47 ms
40 bytes from 10.2.4.76: icmp_seq=2 ttl=64 time=1.44 ms
40 bytes from 10.2.4.76: icmp_seq=3 ttl=64 time=1.31 ms
40 bytes from 10.2.4.76: icmp_seq=4 ttl=64 time=1.67 ms
40 bytes from 10.2.4.76: icmp_seq=5 ttl=64 time=1.63 ms
40 bytes from 10.2.4.76: icmp_seq=6 ttl=64 time=1.76 ms
40 bytes from 10.2.4.76: icmp_seq=7 ttl=64 time=1.78 ms
40 bytes from 10.2.4.76: icmp_seq=8 ttl=64 time=1.40 ms
40 bytes from 10.2.4.76: icmp_seq=9 ttl=64 time=1.31 ms
40 bytes from 10.2.4.76: icmp_seq=10 ttl=64 time=1.62 ms

— 10.2.4.76 ping statistics —
10 packets transmitted, 10 received, 0% packet loss, time 9014ms
rtt min/avg/max/mdev = 1.311/1.542/1.781/0.172 ms 


Again, from what I understand, Ver 4.4 of Firmware is in Private Beta now and should be released in the next few weeks, and it brings QOS and some things like that which are pretty critical for everyone. Then, from there, I understand we’ll see an ePTP mode rewrite shortly after that.

2 Likes

Thanks a lot for your reply.... I really hope that an eptp mode solution will come soon!!!!! Anyway I look forward after 4.4 fw release.... maybe in the end of summer we'll see eptp mode too!!!! Finally!!

Best regards,

Paolo

You are welcome. I agree, we really need ePTP mode in order to be able to use these things wide scale.

That being said, we currently are seeing about 4ms average, plus/minus 2ms, in TDD mode, so thats not horrible for a TDD mode. We do have places where that’s acceptable for our needs. Once ePTP mode is written, they should be spectacular.