Mangolana: improve Solana so Mango runs well

Fund a new company with $600,000 and up to 20,000 SOL (performance based) for one year to employ engineers to contribute to Solana core so that Mango runs with highest reliability and lowest latency. The DAO receives a grant from Solana Foundation of 200,000 SOL per year.

There have been numerous occasions in the recent months of Mango users complaining about reliability issues that fundamentally come from Solana’s growing load. It’s one of those issues, that is very complex to resolve, here are some ideas proposed so far:

  1. Rebuild on another chain, high effort and low uncertainty that same or other issues won’t occur under load. Most other networks don’t experience the same level of demand as Solana.
  2. Fork Solana, create a mango side-chain and don’t allow NFTs. Relatively low effort, but it severely limits the attractiveness of Mango as a trading product. Few of the on-chain composition activity we have seen in the last months would be possible without the ecosystem bootstrapped through a series of hackathons and grants funded by Solana.
  3. Improve Solana, make sure it is resilient against noisy neighbors, high effort and again low uncertainty. First question that comes to mind: Is noisy neighbors even a problem that Solana’s architecture can fix? I’d argue yes.

In case I’m wrong, the attempt at 3) should teach everything needed to excel at 2) so that’s why I am convinced this is the right way to focus our energy in the short-term.

So far many contributors to Mango DAO have worked on Solana core to do quick fixes, well known with positive impact. In order to have people work on this with maximum impact, I need funding to hire the best possible people. This will benefit the entire Solana ecosystem, and allow us to build the kind of public goods infrastructure that will allow Mango to realize it’s promise of being the exchange with the best user experience. It shouldn’t be a 20% effort (which it was for many so far), given how much it impacts the value Mango can provide to its end users.

It’s a weird situation, there is a huge knowledge and opportunity transfer from Mango DAO to Solana here. We give up scarce resources and potentially contribute major improvements to Solana which doesn’t necessarily translate into a guaranteed success for Mango. To address this issue, I applied on behalf of the DAO to Solana Foundation for a grant of 1,000,000 SOL to honor the contributions Mango developers will make under this initiative. The grant will be split in 10 tranches and is contingent on Solana Foundation’s continuous approval based on review of such contributions to Solana’s code base over the coming 5 years. The SOL will be staked with the DAOs validator and become liquid after an 1 year unlock period.

Concrete next steps for this project would be:

  1. Set up a company to pay for salaries and server infrastructure costs
  2. Bring on godmodegalactus full-time, he already contributed multiple patches to mango’s on-chain program
  3. Continue to look for people in my network that have worked on distributed databases before and bring them to work with us full-time
  4. Setup benchmarks to estimate the theoretical workload mango v3 (and future versions of Mango) can execute on Solana’s public testnet
  5. Identify bottlenecks in Solana’s architecture and resolve those to improve both performance and reliability of mango on mainnet. This is especially related to gulf stream transaction forwarding and turbine block propagation.
  6. Operate indexers to speed up latency when accessing on-chain data for mango users
  7. Reduce our running budget for RPC servers to a minimum (currently around $180,000 annually)
  8. Operate a paper trading environment for traders and load tests on devnet / testnet
  9. Work closely with Solana foundation on new milestones to continue distribution of the grant to the DAO

To fund this company, I request $600,000 USDC upfront to finance the first year and authority over 10% of the solana grant once they become unlocked. The USDC needs to be approved before to give economic certainty to the hires of the company. The 10% will be distributed to various contributors to Mango DAO (by no means restricted to people who work in this new company) based on their impact. Progress will be reported on every quarter. Any unused funds will be returned after one year.


This is a pretty mega deal that dwarfs even the fee revenue generated by mango so far. Sounds really great for the long term prospects of solana and mango. Although hopefully the mango DAO doesn’t switch over entirely to becoming a solana outsourcing shop and still manages to succeed with its core mission.

Its core mission obviously being to purchase land and automated farming drones in wyoming. :stuck_out_tongue:


The DJI mini 3 just came out, DAO proposal to buy drones soon…

1 Like

Initial reaction is that this sounds highly +EV even with a much lower SOL price and handicapped for a low probability of success. Also just sounds like a noble cause.

I would be curious to see what the terms and milestones for earning the grant payment(s) look like (ie under what conditions could the foundation refuse to make the initial payment of 200k SOL?)

1 Like

What a noble and ambitious proposal for Mango.
I think MangoDAO should support this proposal.
Even though it’s a big long bet, make a better Solana not just help ourselves, but also contribute the public good to the whole Solana ecosystem.


I believe this partnership makes a lot of sense for both parties. Mango previously had deployed scarce dev resources to Solana for only indirect benefit. This structure enables both sides to benefit from continued collaboration.

However, it would be helpful if a representative from the Solana Foundation could chime in to confirm this is the general understanding. I don’t want to cast doubt on Max’s credibility but at this point the DAO is transferring substantial sums based only on Max’s description of a deal. Some public confirmation of this arrangement and the general structure from the Solana side would be helpful prior to formal DAO approval and fund transfer. @mschneider do you mind passing this request along to your counterparty?

Congrats to Solana and Max for agreeing on a creative structure that should benefit everyone for years to come.


Super interesting project which I would like to work on full time. It will help Mango and the greater community.


Thanks for sharing this @mschneider

Among the options listed there, I believe #3 is the most viable one as Mango and Solana are dependent on each other in a symbiotic way, a relationship which goes well beyond behind Mango being responsible for an extremely high amount of txs.

As next step I believe there has to be more clarity on the exact terms of the transaction, in particular:

  • What is the degree of separation between Mango and the new entity? Will it act as a completely autonomous firm separate from Mango DAO? If so, who will be responsible for its oversight and management of resources?

  • Which entity will receive the proceeds of the grant from the Solana Foundation? The first part of the post mention 200’000 SOL per year, whereas the second mentions 10 tranches for a total of 1M SOL over 5 years with locked SOL to become liquid after 1 year. IMHO these terms could be written more clearly, also: are the tranches of equal size? Are they distributed on a constant time basis (quarterly/yearly/…)? Can external users have eyes into the review process from the Foundation?

  • When you request “authority over 10% of the Solana grant” does that mean authority on up to 100’000 SOL received by the Solana Foundation (out of 1M) OR just 10% out of the initial 200’000 SOL (so the 20’000 SOL mentioned in the first part of the post) for performance/bonus?

  • Last but not least - as @brian_smith_0 has already mentioned, I think it’s a good idea to get the green light from the Foundation too such that we have a fruitful discussion without the risk of putting the cart before the horse.

Deal Structure and incentives

There are three actors in this deal and we need to be sure the incentives are aligned well for everybody.

Mango DAO - If my understanding of the deal is correct, I think this is an extremely good opportunity for the Mango DAO as the economics benefits of the grant are being passed mostly to MNGO holders.

The Foundation is effectively “outsourcing” some of its development needs to the extremely smart guys that created Mango as they master the nitty gritty of solana and can improve it for everybody else (a là public goods). The Foundation will pay the grant to the Mango DAO.

The people behind the new entity are initially old-time contributors of Mango Markets (although new ones will be added), these individuals will receive a compensation made of a fixed sum plus the convex upside coming from their pre-existing holdings of MNGO. The new entity will receive its funding from the Mango DAO.


Overall it seems that the economic incentives of all the three group are aligned significantly, both in the short term as well in the long term. I am positive about this deal/proposal and happy to contribute in any way I can to make it happen.

  1. It’s a new entity that collaborates deeply with the remaining entities participating in the DAO. I’ll personally oversee it.

  2. The grant will be directly going to the DAO, or let’s say as directly as legally possible. It will be part of the on-chain treasury and controlled by token voting through spl-governance.

  3. 10% of the grant (only in case the DAO receives it). I’d propose 10% for the first year as an adequate restriction for now and we can make a follow up proposal for the coming years. So 0, 10,000 or 20,000 SOL depending on success of this venture. :slight_smile:

  4. Solana foundation has the grant already greenlit, this proposal is to get a green light from Mango DAO


Sounds interesting. We’ve seen a lot of contribution to the public good from Mango crew. Can’t take that for granted. We need to make it “granted” from stakeholders.
Solana foundation is actively reaching out to developers these days I think they could be much interested on this too.

Generally curious from the engineering side. Are there identified solutions to the problems (or potential bottlenecks) noted? Or is this entity and funding for pure R&D?

Will the data indexing be for the entire ecosystem? Or is this like a geyser plugin configured purely for Mango related account updates?


Listening to this Firedancer Friday conversation with jump and Solana Labs and thought about the Mangolana initiative.

Curious how this all ties together with respect to improving network performance. Assuming Mango will eventually want to leverage the new client, but will there be any collaboration between the Mangolana and Firedancer teams / overlap in scope?

Any thoughts you can share @mschneider ?

We’ll most likely focus on the solana labs’ rust client in the foreseeable future. The advantage here is that we can have direct impact on Mango’s product quality without waiting for the Firedancer release.

@tomjohn1028 problems are usually latency (read / write). solutions are partially identified and are already in progress. about indexers: this will most likely be decided based on the roadmap of the different products the mango DAO is currently working on. the goal here is to continue publishing our solutions as before under open source licenses.