[Referendum MR50] Moonriver <> Axelar Integration

Governance Democracy Proposals
,
1

TLDR;

  • Moonriver community members have voiced their concerns regarding the lack of a major bridging/GMP protocol being available on Moonriver
  • The Moonbeam Foundation has held discussions with various GMP protocol teams, including Axelar
  • The Axelar Amplifier Pilot Program represents a unique opportunity for Moonriver to be integrated with Axelar’s next generation architecture
  • The costs are not insignificant (~$400K) and it is proposed that funding for the pilot program and operating costs come from the Moonriver Parachain Bond Reserve which currently holds > 390K MOVR (~$9M)
  • The decision to fund and participate in the program is up to the Community, and this post aims to present the opportunity for community feedback

Disclaimer

The Moonbeam Foundation is presenting this opportunity for the community to consider and to decide if pursuing this solution is in the best interest of the ecosystem. As with any GMP/Bridging protocol, there are inherent risks. This post should not be construed as an endorsement by the Moonbeam Foundation.

Background

For the past several months, there’s been feedback from various Moonriver community members that they would really like to see a new GMP (General Message Passing) protocol partner deployed to Moonriver. In discussions with various community members, this was seen as a positive step for the ecosystem, allowing liquidity to be bridged both in and out to other ecosystems such as Ethereum mainnet, Cosmos, etc.

Having heard this feedback, the Moonbeam Foundation has engaged with various teams to gauge their interest in deploying to Moonriver while at the same time trying to understand what kind of support (financial or otherwise) may be needed to make it happen. As with any robust GMP solution, infrastructure must be deployed with appropriate incentive structures in place.

In speaking with the team from Axelar, there is an opportunity for Moonriver to join the Axelar Amplifier Pilot Program. This program seeks to accelerate the transition to a new, more scalable architecture with the goal of connecting thousands of chains.

Summary of Amplifier Pilot Program

Axelar has successfully connected dozens of ecosystems over the past few years and has established itself as a tier-one GMP protocol in the industry.

As more chains are onboarded, changes are needed to solve problems related to scale and complexity. To address this, Axelar is launching Interchain Amplifier – a permissionless model and service that allows anyone to easily and seamlessly create a connection to the Axelar network and all of its interconnected chains powered by the Axelar Virtual Machine.

Axelar is accepting 3 to 5 teams (of which Moonriver would be one) to join a private pilot program beginning in January 2024 with the goal of achieving mainnet launch of Amplifier in Q2.

Participating teams will allocate dedicated engineers that will be responsible for leading the technical development and provide critical feedback during the testing phase through to launch.

The Axelar DevRel and Engineering teams will provide white-glove support to participants, ensuring participant teams are familiar with the Axelar architecture & the components that need to be developed/deployed to successfully connect to the network.

Participation in the program will significantly reduce the amount of time to integrate a chain to the Axelar network. Non-participating teams will still receive support via public channels and monthly calls but will no doubt take longer to integrate.

More details regarding the new Amplifier architecture can be found on the Axelar Community Forum in this post

Forecasted Costs

It is estimated that the first year cost of this initiative will be approximately ~400K USD which covers participation in the Amplifier Pilot Program, Validator Incentives and Relayer services.

This is only an estimate - some costs will become clearer after launch but this should provide the community with a ballpark figure.

Proposed Source of Funding

In order to fund participation in the program and the infrastructure costs for operation, it is proposed to draw upon Moonriver’s parachain bond reserve.

The parachain bond reserve’s purpose is to ensure a parachain slot in perpetuity on Kusama. The current protocol tokenomics are such that the parachain bond reserve is funded via a portion of inflation. Its allocation at genesis was 50,000 MOVR but it now holds more than 390,000 MOVR. Moreover, the Moonbeam Foundation has accumulated a significant amount of KSM to support Moonriver’s parachain slot for some time. (Until Polkadot 2.0 fee markets are introduced, it is difficult to say for how long).

As discussed in this forum post, the Moonbeam Foundation has begun a study to assess network operational costs and the tokenomics of the protocol to ensure long term sustainability of the Moonbeam networks. In an ideal world, this study would be completed, and used as a reference, before committing to reallocating a portion of the parachain bond reserve. However, given current market dynamics, the community may feel that acting now is in the best interest of the ecosystem.

Next Steps

The intent of this post is to present this opportunity to the Moonriver community to solicit feedback.

After community members have had a chance to voice their opinion, ask questions and provide feedback, and if there is enough interest and support, the next step would be to hold a referendum on Moonriver as to whether this opportunity should be pursued and funded as described above.

As the Amplifier program is beginning shortly, the hope is to collect feedback and respond to questions until January 21, 2024 before putting it to a community vote.

11 Likes
2

hey Aaron, thanks for putting it together

Axelar is indeed a great option for Moonriver. especially after the Uniswap Foundation published an extensive Bridge Assessment Report, it recognized Axelar’s strengths, particularly its PoS mechanism and robust technology stack

in fact, I am very interested in whether squidrouter is included in this integration? squid’s cross-chain swap feature is an incredibly important and game-changing for user experience. we definitely need to enhance the user experience, as using the bridge is not so straightforward for most users

could we get an estimate or projection of the costs for this initiative in the second year or beyond?

also, as I understand it, Snowbridge is coming soon to the Kusama ecosystem. this will enable all Kusama parachains to use the Snowbridge for interacting with the Ethereum ecosystem, and possibly extend to others in the future. I believe Snowbridge has already been funded by the Polkadot treasury, suggesting that it might be free to integrate. however, I’m not entirely sure about the specifics behind the scenes. could you please clarify if there are any drawbacks or provide further insights on this matter?

I also believe that LayerZero already supports Moonriver. do we have any estimates on the potential cost of integrating Moonriver with Stargate

4 Likes
3

Hey @aaron.mbf - thanks for the detailled writeup.

Given the existing bridge integration with Wormhole I was wondering if there have also been telks with that team?
I expect the integration to be easier and faster to implement but not sure about the associated costs or Wormholes ability to onboard new projects (I know there’s a pretty filled backlog of onboardings)

4 Likes
4

hey, sik! tbh, I believe that the main reason for using different GMP protocols from Wormhole is because Wormhole assets are primarily utilized on Moonbeam. therefore, diversification on Moonriver is essential. this approach is crucial because if the bridge encounters any issues, both Moonbeam and Moonriver could be adversely affected. It’s about balancing the risks to ensure stability across both chains

4 Likes
5

I think the funds would be well spent, as the impact will be notorious and, if implemented before Snowbridge, it can be specially beneficial to current Moonriver projects not only by the extra users and ecosystems they can tap on but because they can also really use Moonriver as a Canary network for their projects.

Also, being integrated into Squidrouter would be something specially beneficial for the current DeFi ecosystem, (for example, opening the door to other chains assets into DEXes and Lending protocol)

If there’s the possibility to integrate Axelar into Moonriver without joining the Amplifier Pilot Program, how long would you estimate will the delay be compared to joining the program?

And another question pops to my mind, as the participants will need to delegate engineers to take care of this, how this will affect to the work already needed to be done at Moonriver (i.e: Runtime upgrades)?

2 Likes
6

Aside from the reasons @turrizt called out, some discussions were had with Wormhole but given their current priorities, they couldn’t commit to deploying to Moonriver for some time.

2 Likes
7

Great questions as always, @turrizt !

I asked about this and was told that once Axelar is integrated, it is easy for squid to enable an EVM chain into their front-end. However, I’m hoping that someone from the Axelar team will be chiming in on this one. If not, we can try to connect with the squid team to clarify this.

I believe costs for the second year would be in the 200K to 250K range. However, we may be able to do better on relayer services (anyone can run a relayer) so we may be able to get folks to bid for the work, possibly using the treasury). The remaining costs would be incentives for the Axelar validators.

Yes, another option would be to wait for SnowBridge or perhaps even BridgeHub (in which case, liquidity could be brought to Moonriver from Ethereum via Moonbeam using something along the lines of MRL). These options may be more cost effective. However, given the lack of documentation, there remains a lot of uncertainty with these solutions.

We really don’t know when they will be “ready for prime time”. Also, everything will have to go via AssetHub so it’s unclear what the user experience will be like. There may be assumptions around address format that we’d have to work around. There’s also the question of which assets will be supported by Snowbridge/AssetHub. Native USDC and USDT are great but without CCTP (in the case of USDC) or direct withdrawals from exchanges, it’s been very difficult to get liquidity into the ecosystem.

More than likely, once Snowbridge is ready, it will probably take changes to the Moonbeam runtime to get everything working and of course, some sort of front end or SDK work to make use of it.

So in short, I imagine Snowbridge/BridgeHub would be a lower cost option but would take longer to complete and require more effort on the Moonbeam side. I think some in the community feel there is some urgency here but perhaps I’m mistaken on this.

This is something else we can look into if folks think it’s worthwhile. I think I had heard something like 30K USD in incentives per month, so probably comparable.

These are all really great issues you’ve raised for the community to consider and it would be great to hear what others think about the idea and the points you have raised.

7 Likes
8

Going with Axelar makes sense because it is a polished product with a dedicated team working behind it. Even more, their product is Axelar. Plus, the Squid router is an insanely good product that can ensure that liquidity can easily flow into Moonriver.

However, looking into Polkadot<>Kusama bridge with Moonbeam <> Moonriver bridge integration, I agree it makes sense to focus on this for Moonriver, the Moonbeam canary net. Even though they’ve made progress with this, it is unclear when this will be ready as a “polished” product. Nevertheless, it will be going with the bleeding edge of technology and Moonriver is just the place for that. Moreover, there will be no hefty associated integration fees.

So the question for the broader community is: is Moonriver OK with waiting 6-9 months for a polished Moonbeam <> Moonriver bridge to be fully operational? Or should we move forward with Axelar, which will create a bit of a dependency that is difficult to unwind if we as a community decide to move ahead with Polakdot<>Kusama bridge once it is ready?

I don’t have a clear opinion tbh, as I see ups and downs for both sides.

6 Likes
9

hey Aaron, thank you for the detailed answer, it’s much appreciated. IMO, we really need a bridge on Moonriver and fresh liquidity. currently, it’s stuck with multichain assets, and users literally cannot interact freely with the ecosystem because there’s not a lot of liquidity for other parachain assets. so, I think integrating a new bridge to Moonriver is essential. however, the associated costs are a bit high for me. eg, paying 200k annually after the integration seems a bit costly. but it appears that other bridges will also cost roughly the same. If we want to choose a cheaper solution, it would be Snowfork and BridgeHub. I’m not sure if it would be beneficial for the community if someone from the Moonbeam contributors were to explore the Snowfork and BridgeHub options and then share all the perks and drawbacks of such an integration

according to Snowbridge’s docs:

Sending tokens is usually a single step for the user. However, a preliminary registration step is required for tokens which have not previously been bridged

currently, the focus of Snowbirdge is on Ethereum, but this raises questions about potential expansions to other chains in the future

so, the question is, if we as a community skip Axelar and some other bridges just because of the associated costs, will Snowbridge and BridgeHub be worth it? there are probably some drawbacks, such as how much time it will take for the bridge to go fully live, how many different assets it will be possible to transfer, and what the bridge cost will be. I believe I read somewhere that it will be around $20 for one transfer. this could be costly for many users. also, I’m not sure how much time it will take to open channels from other major L2 chains, like Arbitrum, Optimism, Polygon, Avax, etc

another bridge that comes to mind is Hyperbridge, developed by Polytope Labs (Introducing Hyperbridge: An Interoperability Coprocessor) they are already acquiring a parachain slot on Kusama, which looks like an essential option for me. however, we probably should also discuss things with the Polytope Labs team to ask about when it’s planned to officially go live and other technical and financial aspects of such an integration

but yeah, bridges like Axelar, Stargate, which are highly talked about and well-regarded, likely have a significant impact due to their integration with a large number of different chains and their ability to attract diverse liquidity. their cross-chain swap features greatly simplify user experience by eliminating complex bridging things. i believe that our goal and idea are to make things simple, accessible, and incredibly intuitive, so investing in cross-chain swap features seems worthwhile for convenient interaction within the ecosystem

so, I’m probably thinking that although it might be expensive, it’s worth it. also, these bridges seem to be already battle-tested, while Snowbridge and Hyperbridge are just going live. yeah, it probably can’t say anything about security, and a battle-tested bridge could still be vulnerable, but it still gives a bit of confidence. but yeah, I’m not endorsing any bridge, just trying to speak out loud, maybe it will be helpful and share some insights on which option we as a community can choose

5 Likes
10

@leamsi.eth - thanks for your comments and questions.

I think this is a bit of unknown but my understanding is that there is a long queue of other chains looking to get integrated. If I had to hazard a guess, probably sometime in the 2nd half of 2024.

Because we’re EVM compatible, it should be pretty straight forward and shouldn’t really impact core protocol development or dapp development much. The snowbridge/bridgehub options will probably be more demanding in that way but granted, we’ll likely want to support those solutions no matter what. It would just mean we may have to prioritize work toward snowbridge/bridgehub potentially over other initiatives.

However, there will be effort needed more from a devrel side (documentation, verification, etc).

3 Likes
11

I could be wrong, but I think this means is that it will be a single step to get the liquidity as far as AssetHub so there would be some additional work to improve the UX somehow.

Presumably the registration of assets means Polkadot governance, but again, I’m not sure. If anyone in the community knows the answer to this, that would be great but I’ll try to chase this down.

Anyway, I have the same reaction as you: this seems very expensive but at the same time, it’s more of a known quantity and known timeframe than the other options.

4 Likes
12

Once the integration is completed, Squid can easily activate additional EVM chain on their frontend at no extra cost.

9 Likes
13

thanks for clarifying, @marcinpawel, it’s nice to hear that!

4 Likes
14

Hi Aaron, I confirm you, staying continuously in contact with Movr Community… Feeling is that we need a solution… Asap. Waiting additional 6-9 months is not really an option.
So basically I agree with @turrizt, Axelar sounds like a quick and robust Solution, only concern is about associated costs.
Given ofc that these initial 400k are strictly required (and I agree about using parachain reserve for this strategic purpose), I would suggest to focus negotiation about following years costs to be sustained, would be nice to not impact much Treasury funds…

4 Likes
15

Oh, I see you’re having a good discussion here.
It’s great to have so many interested and knowledgeable people sharing their thoughts
:star_struck:

3 Likes
16

I also use squid router quite often and can’t stress enough the importance of Moonriver to be included in the chain roster (Moonbeam is there). It will easily connect Moonriver to EVM chains and enables much more liquidity to Moonwell

2 Likes
17

The price feels a bit steep, but there isn’t a real breakdown so it’s hard to say if it’s reasonable. If the MOVR tokens are just sitting there doing nothing, then that’s one story, but could also be used to incentivize other projects to build on Moonriver. However, that still leaves us without a near-term solution to the problem described.

It’s not clear if the 400k is a discounted price. If this is a pilot program on a new build, there’s bound to be issues, learnings on both sides, and the devrel support from the moonbeam team may be higher (and there’s an opportunity cost in where their time is directed, too). Also, there is some mutual benefit from the shared learnings of the pilot program.|

Based on those considerations, and if this was not a discounted price, I think Axelar should be willing to revisit the pricing.

Also even if as @AlbertoV19 stated it’s difficult to unwind this once it’s in place, at $250k/yr maintenance, it still may be worth it in that any development work could be outsourced and likely pay for itself after a year or two. This could also be brought up with Axelar to negotiate a more agreeable long-term solution wrt pricing.

1 Like
18

Brilliant post and lots of detailed questions and explanations.
With limited info about the costs, we would support this proposal without a doubt.
Looking forward to Q2.

19

Thank you for your work and this great post !
Also, thanks to everyone who has provide opinions !
Personally, I see great potential in this integration !
Axelar has a great team and product, I’ve had experience working with some of their team members!
As for take funds from the parachain reserve I also agree take them !

3 Likes
20

A very informative post, a lot of work, thank you!

Based on the explanatory conversation I have re-read, I also see certain risks associated with uncertain costs and interactions that should be brought from beginning to end. However, the positive result we can observe (if the task is completed) exceeds the risks.
I also interacted with the Axelar team, they can finish the job, and I have confidence in them.
This decision will benefit you!
So I’m positive and supportive!

2 Likes