pfsense latency test

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

T_Minus

Build. Break. Fix. Repeat
Feb 15, 2015
7,649
2,065
113
Are there any specific, repeatable tests to get an idea of latency?

I'm curious how different CPUs affect latency for pfsense... obviously it's drastic for CPU intensive activities, I'm talking about basic stuff. I think there's other CPU off-loading that would be good to take into account for performance.

Has anyone done this or seen any graphs? Not expecting anything huge just curious more than anything.

I noticed pfsense appliances (upper end) are XeonD and Atoms but they're all 2-2.4ghz what would 3.7ghz do? Anything?
 

nitrobass24

Moderator
Dec 26, 2010
1,087
131
63
TX
Look at bufferbloat.net - Bufferbloat is the undesirable latency that comes from a router or other network equipment buffering too much data

They have a shell script that basically saturates router bandwidth using netperf to generate traffic while simultaneously running a ping test. I find this particularly useful when comparing configuration changes on the same hardware (e.g. you can see the impact turning on DPI, or IPS type functionality can have on latency and throughput).
 

Joshh

Member
Feb 28, 2017
61
16
8
43
Look at bufferbloat.net - Bufferbloat is the undesirable latency that comes from a router or other network equipment buffering too much data

They have a shell script that basically saturates router bandwidth using netperf to generate traffic while simultaneously running a ping test. I find this particularly useful when comparing configuration changes on the same hardware (e.g. you can see the impact turning on DPI, or IPS type functionality can have on latency and throughput).
Pretty cool, will have to give this a run. I have wondered the impact of things like IPS etc, although I just assume things are fine as speeds are never affected and I always have had decent hardware to throw at Pfsense.
 

T_Minus

Build. Break. Fix. Repeat
Feb 15, 2015
7,649
2,065
113
@nitrobass24 awesome suggestion, thank you! I've never heard of that until now.

My hunch is as long as you're not running a ton of packages/add-ons 1ghz vs 4ghz shouldn't really matter too much unless you're trying to do some huge bandwidth or something crazy :D