[PROP #2][APPROVED] [UPDATED] Increase the Active Validator Set to 125

Dymocracy
1

Proposal: Increase the Active Validator Set to 125

Dymension continues to make strides as the Home of the RollApps within the Cosmos ecosystem and beyond, evolving in tandem with the burgeoning activity on the chain. With the successful launch of the Dymension chain and the increase max_bytes and max_gas in BlockParams, it is evident that our ecosystem is expanding rapidly. In light of this growth, we propose to further enhance the infrastructure by increasing the number of active validators from the current 100 to 115.

Background: The Dymension commenced with an active set of 100 validators, that left some proficient validators, supporting the protocol from scratch, out of active set. Since the launch, the chain has demonstrated stability and remained incident-free. However, the Dymension validator community has experienced substantial growth, resulting in a considerably high minimum stake requirement for active validators.

Notably, other Cosmos-based chains, including Osmosis and Juno, have successfully raised their validator caps to 150. This illustrates the feasibility and positive impact of such parameter upgrades.

What do we propose? This proposal advocates for an increase in the number of active validators to 115. By doing so, we aim to provide smaller and newer validators with an opportunity to be actively involved in the Dymension ecosystem growth. This adjustment will not only foster a more inclusive validator community but will also encourage greater delegation from the broader community.

As a dedicated validator ourselves, jayjay Validator believes that the continued growth of the validator ecosystem is pivotal for fostering collaboration and overall network expansion.

Community Input: We invite feedback from the community on this proposed adjustment. Do you support increasing the validator set size to 115? If not, we are eager to hear any concerns or considerations you may have against this upgrade. Your insights are invaluable in shaping the future trajectory of the Dymension.

[UPDATED]

After a long discussion with all regardless members of the Dymension ecosystem and taking into account all expressed thoughts and concerns, we are eager to update our proposal and expand the Validator Active Set to 125.

Pros:

  • From Validator Rank 116 to 125 - are still 2.7M $DYM delegations in the inactive set and not earning any rewards for delegators/validators.

  • There is neglect-able negative impact expending the set straight to 125 Validators.

  • Among the Validators in the inactive set are many great community validators which couldn’t make it in against major whales, such as: Cosmospug, Stakelab, Dora, Danku_zone, Oni Validator, SmartStake, we hope to encourage them to further contribute

  • Sooner or later the set needs to be expanded again anyway and onboarding the community validators already could align them right away. The communities for those Validators could be reactivated also in respect of the upcoming news, potential airdrops, delegations and many more…

  • Having 150 Validators long term is a good target and makes sense - we can also see on other networks that bigger sets are working fine and are quite normal for cosmos chains.

  • We’ve contacted with validators in the inactive set, and all of them have stated to maintain their infrastructure and are looking forward to join the set after a proposal approval.

Cons:

-Based on immaturity of the network, awareness about the chain behavior in the future when rollups will be deployed, which may lead to additional load (cpu, ram, network, disk).

76 Likes
2

Would be interesting to cover the pros and cons of increasing the validator set with more signers. Knowledge can be very powerful.

21 Likes
3

This proposal is a no brainer.
Dymension is definitely attractive enough to gather a larger number of reliable validators. I was surprised by the initial number of 100 which from the start lead to quite a few hesitations from many holders who wanted to stake with “smaller” validators but weren’t sure these were going to stay in the top 100 long term, which lead to more centralization around the top validators than we could have hoped.
115 is a good number to move towards 150 on a longer term horizon.

29 Likes
4

In the inactive set, there is approximately $5.8 million DYM among the first 15 validators, meaning a significant number of delegators and validators with a total stake of $40 million cannot earn APR income and contribute to the security of the network. Currently, there is around 500,000 $DYM on the 100th validator, equivalent to about $4 million. Entering the active set is currently quite challenging, and the current VP distribution still poses a security issue for the network, as the top 4 validators control more than 33% of the total voting power in the network. Therefore, we support this proposal.

25 Likes
5

For this range of validators there should not be a tangible impact on the networks performance. Dymension blockchain does not aim for fast performance, as that is delegated to the RollApp layer.

Given that such a large stake is sitting outside of the active set and can contribute to the security of the protocol I believe it makes sense to support this prop.

25 Likes
6

This proposal is certainly important for the decentralization of the network. Because users prefer large validators to avoid APR revenue stagnation and to avoid missing out on possible airdrops. This leads to the concentration of voting power in a few hands.

9 Likes
7

The NODEJUMPER team stands behind this proposal. We actively participated in multiple testnets, diligently contributing to the network during its early stages. While we, along with a dozen of other proactive validators, initially secured delegation from the foundation, the entry of major stakeholders with substantial investments led to our exclusion.

This situation results in several consequences:

  1. The efforts of early adopters are undervalued, potentially diminishing motivation for individuals to strive for excellence within the network.

  2. Foundation delegations lose their efficacy as funds are tied to inactive validators, a prevalent issue.

  3. This scenario adds pressure on token prices. Large self-stakers can now freely sell their rewards since there’s no incentive to increase self-stake through restaking for eligibility in subsequent delegation rounds, given it won’t help to move to the active set.

20 Likes
8

Suggestion: Expending the Validator set straight to 125

Reasoning:

  • From Validator Rank 116 to 125 - are still 2.7M $DYM delegations in the inactive Set.
  • There is no relevant negative impact expending the set straight to 125 Validators.
  • Among the Validators in the inactive set are many great community validators which couldn’t make it in against major whales, such as: Cosmospug, Stakelab, Dora, Danku_zone, Oni Validator, SmartStake.
  • Sooner or later the set need to be expanded again anyway and onboarding the community validators already could align them right away. The communities for those Validators could be reactivated also in respect of the upcoming news, potential airdrops, delegatiosn and many more…
  • Having 150 Validators long term is a good target and makes sense - we can also see on other networks that bigger sets are working fine and are quite normal for cosmos chains.

The be transparent we as Stake&Relax are also among the Validators in the Ranks between 116 - 125 and would love to contribute to the Network.

24 Likes
9

I’m convinced that this proposal comes too soon.

The chain has only been running for 10 days and you’re already talking about increasing the number of validators.

I know it’s very frustrating for a validator not to have been able to keep his validator in the active set, and most of us are still working every day to keep it there.

Before we think about increasing the number of validators, perhaps we should focus on resolving the remaining problems after launch and on deploying some roll-apps to give this chain some activity.

Friendly, a fellow validator
Crosnest

12 Likes
10

Thank you David for sharing your concerns with us.

The network and consensus layer itself are being operated smoothly and showed that it’s able to handle with big amount of transactions during first days of the chain launch and continues to do so.

Inactive validators operate own infrastructure and contribute to the network from technical side as well as active ones, but one of the most important disadvantage of being inactive is inability to retain loyal stakers and attract ones.

11 Likes
11

VNBnode will vote YES for this proposal to increase to 125 validators.
Alot of delegation from foundation is out of active top 100.
For more decentralization and security, the 115-125 is a good numbers for the first period of chain.
i have only a wonder that: when this proposal can be on-chain for approval. If this proposal takes longer time to get approval, it will be more difficulty for smaller validators.

6 Likes
12

My question is, is approving this proposal now a necessary step for the network at this point, or do we want to wait until specific optimization criteria have been met and ensure that certain network thresholds have been reached first?

On another note, what happens when inactive validators get into the active set with this proposal and then get removed again? Maybe consolidation would be something to factor in too.

7 Likes
13

Hey David,

It’s great to see diverse opinions here.

We think that the issues on the network and lack of active rollapps are not direclty connected to the number of validators. Also, the contribution of $DYM’s outside the set to active security will be beneficial. Furthermore it would be a step forward towards decentralizing the network more, seeing the current balance problem.

Best Regards

7 Likes
14

In fact, it could be related.
without roll-apps, the chain is almost inactive, it just piles up blocks.
If roll-apps are added, the chain, and therefore the validators, will have to take on the additional load (cpu, ram, network, disk).

More validators implies a more complex consensus and an increase in the flow of packets on the P2P network, causing a greater cpu, ram and network load, in addition to those required for roll-apps.
I agree that this increase may seem negligible, but it can’t be ignored.

9 Likes
15

For my opinion, when a validator was selected for delegation program by the foundation, that means they met the requirements and criteria and for sure all the selected ones in delegation program more or less contributed for the project since devnet and testnet phases and also supported for developing the communities.
But may be the delegation is not enough to keep them in top 100 due to the high delegation from big hands to some validators.
So it would be wasted for the delegation from the team and the selected ones could not contribute more on their role. The voting power is centralized with only few top validators.
So for the proposal, it could be fair for both foundation’s selection and also selected validators that all the selected validators in delegation program should be in active set for at least 6-8 months, the delegation from big hands should support for those selected validators as well.
As long as, those validators are still willing to contribute and meet the requirements for validators, they will have chance to contribute and further support on their role.
I do hope, this can make some senses for foundation and all active and deactive validators.
We should together make this proposal become true.
An alternative way of supporting, The foundation could do a list of selected validators in delegation program, then big hands could know the name and consider for their delegation or supporting for the selected ones.

6 Likes
16

Thanks for making this post!

I have to agree with @David_Crosnest here, this sounds ideal but it feel too early to increase the set this much so early. We attempted to get into the active set on launch but unfortunately could not make it, nor were we chosen to receive a foundation delegation despite our relaying and endpoints for both testnet and mainnet. So of course we’d love the set to expand to have another chance, but taking a step back it seems this is being rushed.

What is the actual goal here besides just having a larger active set? That amount of DYM will have a negligable effect on consensus and can start to earn rewards by redelegating to an active validator. This won’t really make any difference on the decentralization front, that will take other efforts. Is there something else I’m missing that would remove the top heavy stake problem? @Huginn

The Dymension team has already given out foundation delegations for the next 8 months with no changes I believe, if anything the expansion should just cover the teams given those delegations or the foundation should decide to remove/increase those delegations to reflect the proper active set. Tagging @Shaolin. For the largest positive impact and if the goal is really to add more good validators to the active set, it would be best to wait for an expansion that is coordinated with foundation delegations and the Stride council delegation. Otherwise, this is too early for an expansion of 25% of the set.

7 Likes
17

The actual goal is to strengthen the security. As @Shaolin said, Dymension chain does not aim for fast performance, as the voting power of 15 validators is not at a level that would noticeably reduce the performance anyway. Additionally, since most of the delegations in the inactive set comes from the foundation, they are unlikely to be redelegated, so that inactive validators can continue to claim the ability to enter the active set. For a healthy active set, validators at the top of the inactive set should be strong and assertive in their claims to be included in the active set, even if they are not currently included.

6 Likes
18

A similar proposal was raised on celestia forums recently: Increase the Validator Set by 5-10 slots - General - Celestia Forum

I would love to see larger sets in cosmos generally, but my concern is that the number seems arbitrary with little discussion around the drawbacks of a larger set.

I don’t know myself but am curious: at what point will performance decrease substantially and require better hardware, and what exactly is the bottleneck or concern that chains often have with increasing sets - tendermint val p2p propagation times, or complexity coming from more validator communication ? Are there any stats from other chains where the set became too large, is it worth testing?

Some numbers on what is sustainable would also guide future set expansions. For example is 200 too many validators?

5 Likes
19

I am not sure why someone think that it’s “too early” for some “necessary”. Let’s take a look at PROS and CONS of this proposal.
PROS:

  1. More decentralization: Of course by adding more validators, together with their communities will bring not only more decentralization but also more users to the ecosystems. Most of validators from top 101 - 125 are from professional validators/company with their own communities. They could bring more than just “voting” to the ecosystem.
  2. More securities: With more validators, the network becomes more resilient to attacks or failures.
  3. Efficient utilization of fund: The fund from foundation’s delegation and from all those validators could be efficient utilized for “directly” contribute in the network in a more decentralization way than redelegation to active set. If all people in the discussion think that finally the expansion is need, so why don’t we make it sooner instead of later? and why should foundation has to withdraw the delegation and then make another selection for delegation at another time later? it does not make sense and benefit.
  4. Enhanced Validator Competition: Increasing the number of validators can encourage healthy competition among validators. This competition can incentivize validators to provide better services, maintain high uptime, and offer competitive fees, ultimately benefiting token holders and network users. With the current situation, the gap of delegation between top 100 and top 105 are very large, then active validators are safe enough to stay there. There is no competition for better services or even the commission rate. With 115 or 125, sure there is a continuously competition between at least 30% of validators to become better and that will contribute more for the network.

CONS:

  1. More complexity for consensus: As theory, yes there will be more complexity for the consensus of more validators, but with the aim of chain, no high through put at mainchain, That will be done at rollapps level. The more securities and network resilience, increasing to these size will bring more benefit than the negative.
  2. Higher resource required: At the moment the minimum hardware of 4 cores, 8 GB RAM and with the resource from most of professional validators, currently managing many much larger resources for other network, the concern of increasing resource is negligible. For near future, the work load of validators are still can be managed with the current minimum resource.
  3. Low stake barrier: With the current price of DYM and the amount of minimum staking among top 120, it’s not a low barrier for anyone to join freely. Then it’s still good to the purpose of only selected one can join and participate as validator. So again this is not an issue.
    If you have any other PROS and CONS, please discuss about it. With this consideration, i think it’s better to support this proposal at sooner stage than at later stage.
4 Likes
20

Finally it became on-chain voting: Dymension Portal App
Thank team and foundation to listen to our voice.
The result of voting will tell us what is the best for network.

4 Likes