The ultimate goal of nomination pools is to lift limits of pool members and number of pools on the network. To that end, I’m proposing the following limit increases:
Polkadot
maxPools: 256 → 2,048
maxPoolMembers: 16,384 → None
maxPoolMembersPerPool: None → None (not changed)
Kusama
maxPools: None → None (not changed)
maxPoolMembers: 65,536 → None
maxPoolMembersPerPool: 1,024 → None
Westend
maxPools: 512 → None
maxPoolMembers: 524,288 → None
maxPoolMembersPerPool: None → None (not changed)
If the community agree on these figures I will create a proposal.
Max pools on Polkadot is proposed to increase 8x to 2048, removing the perception of them being limited & promoting more interest from the community. Kusama already has a None limit on max pools.
We are now reaching very high numbers for maxPoolMembers on both Kusama and Westend, both of which are probably not impactful now - both from an economic point of view and a safety precaution in the rollout phase. None values in place of these numbers would be less ambiguous - we should not try to give meaning to these very large numbers if they do not contribute any benefit to the network.
maxPoolMembersPerPool at None is something that is already in effect on Westend and Polkdaot. This proposal will also update Kusama & remove the limitation.
Is there a bottleneck with the current maxPools and maxPoolMembers or what is the reason for the change?
I dont see how it is useful to have unlimited maxPoolMembers just because it is possible.
Conversely, we should not set these large numbers and pretend there is meaning to them, whereby the larger they become the less (if any) impact they have on the network. These numbers, while useful in the rollout stages, do not carry much meaning now, & carry more ambiguity than a None value would.
Polkadot is almost halfway through the 16k safety cap and I given the long duration to enact changes, I think it is wise to be proactive and initiate the discussion ahead of time. So I am in favor of having this discussion now rather than when it is already almost too late.
It would be useful if you mention all 3 relevant numbers in all 3 networks for better comparison:
maxPools
maxPoolMembers
maxPoolMembersPerPool
As for the measures, what I suggest is:
Kusama and Westend should be, as suggested, set to unlimited number of pools and members across the board.
For Polkadot, I would still suggest keeping some bounds on maxPoolMembers and maxPools for safety, but once a few more quarters are passed with no issues, all of these metrics could be set to None (ie unlimited) unless if strategically decided otherwise because of a specific reason.
Lastly, the same proposal could also be used to revisit the commission parameters, if need be.
Excellent shout with commission updates. I am leaning on the side of caution given the issues that Westend presented to us recently. Perhaps for now we can leave Polkadot alone, and closely monitor / test on Kusama - where GlobalMaxCommission is 10%.
If this discussion / proposal requires more time then these thoughts assume and we manage to carry out commission testing prior to this proposal going live, then we could open the door for combining.
I have opened a new issue to keep track of increasing the limits in the nomination pool limits and other staking related configurations. I listed the nomination members pool configurations and current state for Polkadot, Kusama and Westend.
In summary, I agree we should remove all the pool limits in Kusama and Westend. As for Polkadot, I suggest the following:
maxPools from 256 -> 512 (currently at 128 pools)
maxPoolMembers from 16,384 -> 49,152 (currently at 11,250 members)
maxPoolMembersPerPool keep as no limit (None)
If we agree with these new limits, I’m happy to prepare the gvt proposals.
Suggestions all look good to me. Perhaps we should just be aware that with having the limits on Polkadot, we probably have to revisit them in a few months time but I think it is fine to do one more slow progression and then set them to super large values or None.
