Cambium... Question regarding PMP Proportional Scheduler

Coming from Legacy, we’re looking to rework some things to try to improve overall performance and penalize poor installs over clean installs to allow the overall network performance to improve as we fix clients with poor signal.

Do the previous Cambium-Canopy-DLBR/ULBR and Cambium-Canopy-DLMB/ULMB conflict with the new plan attributes? or can we leave those in place in our radius responses? In case we decide to roll back to legacy.

Trying to understand the difference between enabled vs enabled with threshold, i’ve read the docs but little bit confused.

Does the traffic get impacted when the frame is not congested? If we’re running at 50% frame utilization, and a customer is locked for 8x but is at 2x will they still be frame limited to the reduced capacity?

If I set all radios to 8X Locked it will allocate everyone 8X slots equally so technically we should get max 20/40mhz throughput if we’re at 100% frame utilization because we’re passing 8x sized space for every part of the frame even if some of those customers are drastically reduced throughput because of poor signal, correct?

I take it that 8X locked and 8X threshold is the same? Everyone’s locked at 8x frame proportions for their package sizes?

I’m failing to visualize however the difference between the Locked vs Threshold configs at lower comparative settings, it would seem that threshold would leave higher modulations alone, while locked would impact people even more at lower modulations, I guess i’m just failing to see why to use Locked ever over Threshold?

Proportional scheduling does not impact traffic when the frame is not congested.
All traffic gets through.

The Legacy burst rates are not used if proportional scheduling is enabled.

Locked modulation calculates a users slot allocation based on the selected locked modulation, not the actual modulation.

Locked below threshold calculates a uses slot allocation based on the acutual modulation unless the modulation is below the selected threshold, in which case the SMs allocation will be based on the selected threshold.

For example,
An SM with a locked modulation of 6x:
running at 8x will get slots as if it was running on 6x (so this would be above its plan)
running at 6x will get slots based on 6x (at its plan)
running at < 6x, will get slots based on 6x (below its plan)
An SM with a locked modulation below threshold of 6x:
running at 8x will get slots as if it was running on 8x (at its plan)
running at 6x will get slots based on 6x (at its plan)
running at < 6x, will get slots based on 6x (below its plan)

The idea behind locked with threshold is to prevent a lower performing SM from taking more slots to achieve its plan and thus preventing it from negatively affecting plans on all other SMs in the sector.

1 Like

Ok that seems to be about as i expected with the same on locked or threshold when set to 8x frame allocations, the main difference is only for users above the settings allocations.

I think my confusion is that last sentence of yours because both locked and locked with threshold seem to prevent low modulation SMs from eating up bandwidth from the overall sector the way I read it.

In both cases a QPSK SM can only get as much as if they had been at 64QAM worth of frame space allocated so instead of getting 12mbps and eating the entire frame he might only get 1mbps, the locked seems to not just restrict the poor modulation SMs it actually gives bonus bandwidth allocations to the higher modulations even in excess of what they would be allocated in threshold.

I’m having a hard time seeing any use case where locked, would be the right choice, unless you really wanted your lower modulations to get even less bandwidth because your 8X guys are taking up even more of your frames real estate than they otherwise would have.

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.