Blockchains not only need to be technically good. Besides the protocol and implementation levels, a key element in the success of a decentralized system is having different and independent groups using it, operating it, and governing it. The economic incentives in decentralized systems to achieve such participation by all these different groups have gained attention from researchers who are now interested in “tokenomics”, as a new field of study.
In this article, we are going to explore Solana economics, focusing on the stimulus to network node operators, or validators. We conducted an analysis of the inflation model, the costs and rewards to validators and stakers, as well as the current network activity levels. We also estimate the minimum stake required of a validator in order to break-even, and estimate the impact of different market scenarios, considering the most important variables and how they affect validator profitability.
For this purpose, we built the Solana Validator Dashboard and the Solana Validation Cost Estimator.
The Solana inflation design has defined SOL emissions as starting at 8%, and decreasing by 15% every year. The model was activated on February 10th, 2021 with the payment of 213,841 SOL.
As of July 2022, Solana’s inflation rate is around 6.8%. The staking yield is equivalent to 9.1%, as 75% of the total supply is currently staked (i.e. total inflation rewards are distributed to staked tokens only, resulting in a dilution of non-staked tokens). The rate does not reflect the yearly emission rate. It can be considered a target instead, and the mechanism behind it is broken down below.
Solana’s inflation model considers 400ms block times even though it is mentioned on Docs that the current implementation targets block times to 800ms. The recent average is around 650ms but with high variance.
Although Solana remains extremely performant to the everyday user, the difference in slot times directly impacts the economics and business viability of running a validator on Solana. Longer block times will result in smaller rewards, given a smaller number of epochs in a calendar year, decreasing the amount of SOL distributed to network participants.
In every epoch, Solana calculates the number of tokens instantiated for the inflation pool. The result will be the amount of SOL tokens to be distributed to validators and stakers as inflation rewards, according to the voting and staking status from the previous epoch. 0.45 SOL is the approximate amount currently allocated and distributed among eligible validators in each slot — 195 thousand SOL per epoch.
Block times impact inflation rewards as the function will taper the initial rate (8%) given how many slots have passed since inflation activation on Mainnet — as a proportion of how many slots fit in one year.
Considering an average block time of 650 ms, the inflation being distributed in every epoch is equivalent to a 4.1% yearly rate and the stake yield falls to 5.5%, instead of the 6.8% and 9.1% previously assumed.
Also relevant to validator economics will be the commission. In fact, stake owners, a.k.a. delegators, earn the inflation rewards. Validators earn a portion of it represented by the commission. In the plot below, we can see that a common fee for public nodes is around 10%. There are only 81 nodes charging a 5% fee or smaller. 100% commission is assumed to refer to private nodes (100 validators).
Block reward from transaction fees varies according to network activity. Recent average is around 0.01 SOL per slot. Total per epoch increases with voting power, as the number of slots attributed to the validator is based on the proportional stake.
Theoretically, as inflation decreases with time, validators’ rewards would be supplemented by the increase in transaction fees. The assumption can eventually become a truth as the network matures. Some plots below show that currently this is an unfair assumption:
1- The market’s cyclic nature — the number of non-vote transactions will not necessarily be growing over time. Total transactions (vote + non-vote) picks in Oct21, around 180 thousand in one day. And falls to less than 100 thousand transactions in Apr22.
2- Solana network has invested in growing the network of validators. The plot below shows the number of unique rewards recipients (addresses).
3- As a consequence of voting power dilution and lower network activity, rewards obtained from transaction fees decreased for validators in an individual perspective.
We split the cost into i) hardware, colocation, and bandwidth, to host the validator and ii) personnel, which can vary significantly. The official recommendations can be found on the Solana Documentation.
The vote is an affirmation that a block it has received has been verified, as well as a promise not to vote for a conflicting block. — Solana Docs
Validators are expected to vote on the validity of the state proposed by the slot leader. A validator node, at startup, creates a new vote account and registers it in the network. On every new block, the validator submits a new vote transaction and pays the transaction fee (0.000005 SOL).
Validators usually own (a portion or the total of) the staked tokens, a.k.a. self-staking. In this case, the cost of tokens depends on the average price of acquisition. For the purpose of the current analyses, we will consider the validators only to own 100 SOL at a US$ 50 price.
The Solana Foundation promotes the growth of the validator set through the Solana Delegation Program. Applications require small validators to achieve the “baseline” criteria, which includes running a node also on the Testnet, in order to receive 25,000 SOL. Those who meet the baseline criteria and also the “bonus” criteria can receive an extra (dynamic) amount in the delegation. A recent post on stake delegation strategies and why delegation programs are needed, goals, and criteria can be found in How can networks nurture decentralization?
In summary, Solana validator’s profitability depends on the current inflation rate, block times — reflected on the number of epochs in one year, the voting power, the total supply, the number of transactions, the cost structure, and the SOL market price.
For the three operational levels stated above, we will look at three different economic scenarios: optimistic, average, and pessimistic, with the average scenario being the closest to the current values.
The average market price in one year is fixed at $50 for the purpose of break-even analysis. Different price scenarios can be evaluated in a further session.
We found that the 40,000 SOL to break even would be a realistic amount for a small validator, on an average scenario, close to current levels. The number grows to 253,000 SOL for the medium setup. A professional validator would need more than 1.3 million SOL staked.
For a validator with a 0.01% stake, we estimate a 25 SOL reward from transaction fees in one year. The voting process costs around 200 SOLs per year for each node operator. Therefore, small validators are dependent on inflation rewards to achieve break-even, and ideally, become profitable. Around 350 thousand SOL staked would be needed to fully cover the voting cost, when considering rewards from transaction fees only.
Although the number of validators may be considered high compared to other Proof of Stake networks, 71 accounts are responsible for 57% of the total 365 million SOL staked.
The majority of validators currently stake between 80 and 90 thousand SOL, as seen in the plot below. There are at least 138 (7%) instances of the validator client with stake amounts smaller than 40 thousand SOL, the estimated break-even level for a small validator.
Simulation shows that medium and professional validators are more sensible to fluctuations in the SOL market price than small-size validators. Considering SOL average price in a year to be $75, the break-even level decreased by more than 30% for medium and professional levels and only 7% for small validators. A similar effect is found if the average price drops to $25.
Solana validators and stakers have seen rewards decreasing with higher block times compared to the projected rewards from the initial inflation model. As additional factors, the network experienced a contraction in non-vote transactions during the latest months and the expansion of the validator set.
According to the break-even levels discussed above, an 8.85% inflation target would be the rate level to reflect an effective 5.5% emission in one year, considering 650 ms block times (6.3% if 550 ms block times). Assuming 75% of total supply is delegated to validators, staking yield would become 7.1% in one year and the minimum amount in stake to break even drops by 24%, to 32 thousand SOL.
The inflation rate is even more relevant for small validators’ profitability, compared to transaction fee rewards. Adjusting the inflation model according to the actual network configuration would reinforce the interest of those validators staking less than 40 thousand SOL. Supposing the 8.85% rate simulation above, approximately 21 more validators (1.12%) would reach the break-even level — that is the number of validators currently in range 30-40 thousand SOL in stake.
In this study, we explored the variables behind the Solana validator economics, estimating profitability levels for different market scenarios.
Fee markets are now live on Solana but the adoption of the priority fee by dApps and general users at the moment is low, with the proportion of around 4% of transactions paying a higher fee than the fixed rate. It has been in an uptrend since launched, in late July.
Go to the Solana Validator Cost estimator in getguesstimate to explore the relevant variables, their interactions, and correlations. Thanks, Ruud, Chorus One engineer, for building it.
“Look below the surface and you will find that all seemingly solo acts are really team efforts.” —John C. Maxwell
This article was brought to you by Chorus One. With meticulous review by Felix Lutsch and Ruud.
Chorus One is one of the largest staking providers globally. We provide node infrastructure and closely work with over 30 Proof-of-Stake networks.
Website: https://chorus.one
Twitter: https://twitter.com/chorusone
Telegram: https://t.me/chorusone
Newsletter: https://substack.chorusone.com
YouTube: https://www.youtube.com/c/ChorusOne
Maximum Extractable Value (MEV) represents a fundamental concept in cryptoeconomics, highly affecting permissionless blockchains. MEV is the consequence of the design of protocols and brings with it bad and good externalities. Indeed, not all MEV can be considered benign as some represent an invisible tax on the user, e.g. check out one of our previous articles — Solana MEV Outlook. In general, MEV can also be an incentive for consensus instability, see e.g. the time bandit attack. However, considering all types of MEV as bad externalities is wrong. There exist benign forms of MEV that ensure protocol efficiency, and one prominent example is arbitrage. Let’s imagine that some user swaps a huge amount of token A on a specific AMM (huge with respect to the total amount in the pool) and that this transaction creates a $5,000 arbitrage opportunity. All users that swap tokens in the same pool and same direction will see their output lowered with respect to the actual value. Thus, whoever exploits this MEV opportunity will also bring the market back to parity with the true price. This will make the AMM more efficient without harming its users in the process.
On Solana, MEV still represents a dark forest since no one has pointed a flashlight at it. This is because Solana is a much younger blockchain compared to Ethereum, which can be seen in the lack of products like Flashbots. One project that is moving in this direction is Jito Labs, which recently delivered the first MEV Dashboard for Solana representing an explorer aimed at illuminating MEV — see here for an introduction. However, it is not the only one trying to fulfill this duty. Pointing lights on some Solana Decentralized Exchanges (DEXs) in order to illuminate the dark forest is one of the key objectives at Chorus One. MEV is a consequence that will be a crucial factor for the future of PoS networks and we are continually looking for the best way to ride it. You can explore our Solana MEV dashboard here.
It is important to understand that a simple copy of Flashbots may not be good for Solana, since it represents a drastically different network from Ethereum — and Jito seems to be something intrinsically different. In this article, we are going to assess what are the MEV challenges Solana faces. We’ll also review the status of our internal research regarding MEV.
In Section 2, we’ll analyze the current and future status of MEV on Solana, with a detailed analysis of what we found on-chain in Section 2.1.
In Section 3, we’ll discuss some implications of the current MEV strategies and how these can affect the functionality of a PoS network.
MEV has a specific supply chain, which “describes the chain of activity which helps users transform intentions into finalized state transitions in the presence of MEV ”. However, despite this “universal” definition, MEV on Proof of Stake (PoS) networks is drastically different from what it represents on Proof of Work (PoW) networks. This is for several reasons. For sure, the most important difference relies on the possibility of knowing for sure that a validator will propose a block at some point. Further, validators have delegators and can offer to them a portion of the MEV revenue (e.g. lowering the commission) attracting users to delegate with them. This makes MEV on PoS networks a growing business model, which constitutes one of the building blocks for cryptoeconomic incentives. From one side, we have validators who can use MEV revenue to reduce commission rate — even go to negative values — by returning all incomes to the delegators. On the other side, we have incentives for Layer-1 (L1) blockchains to improve network performance. This is because, if the “scaling problem” is solved by the introduction of L2s, the MEV and transaction (txs) fees are also moved away from the main chain, weakening the L1 business model.
This is exactly what blockchains like Ethereum are facing right now, representing one of the great risks over the next few years. See this Twitter thread for a better understanding of the topic.
But, what is the current status of MEV on Solana? Let’s start from the beginning. Solana does not have a public mempool, meaning that some bad externalities of MEV are very difficult to achieve. However, Solana is not free from them since MEV extraction may produce a bad performance of the network, e.g. spam txs, dropped txs, etc. Indeed, some MEV opportunities only exist if searchers run their own validator, inspect txs that come to them, and run an MEV-extraction code on top of it. Having a high stake and getting access to more MEV opportunities is not an easy task. This dramatically reduces the likelihood of being highly profitable, as the distribution of MEV revenues averages around zero, with a tail towards higher values — see Fig. 2.2.
Note that this is obtained in a specific time window, so it is only representative of the shape of the actual distribution.
Since txs fees on Solana are low and MEV opportunities can bring validators more profit, validators are incentivized to auction off their block space to searchers, or at least some rumors are pointing towards this possibility.
Further, on Solana, fees are currently fixed and cheap, meaning that if there is high competition in a specific market, users face the risk of not getting transactions executed. Since a gas-fee auction is still missing, currently MEV searchers spam transactions to the leader (and following validators in the leader schedule) in the hopes of “winning the battle”.
Lastly, on Solana, MEV competition may incentivize validators to perform denial of service (DoS) attacks on other validators in order to leave the spotted MEV opportunities just there, sitting on the table where they are until the attacker can extract them.
The current status of MEV indicates how bad the problem of blockspace-waste is, which resulted in degraded performance for normal users. At the time of writing, according to what can be found on Jito’s MEV dashboard, we have 12,072,328 successful arbitrages against the 350,179,786 unsuccessful ones in 6 months (i.e. a 3.3% of success rate). If we also include liquidations, the success rate goes down to roughly 3%. The total extracted “good” MEV is around $33M. Of course, this is only a lower-bound since MEV can be created any time a user interacts with a blockchain, and smart contracts enable a functionally infinite number of potential interactions. Thus, it is computationally infeasible to calculate a blockchain’s total potential MEV by brute force. Further, we have some previous analyses that show how a huge amount was extracted during periods of stressful market conditions, e.g. $13M MEV during Wormhole Incident and $43M Total MEV from Luna/ UST Collapse on Solana.
Future Solana improvements aim to introduce several features, forcing current MEV strategies to change. Introducing these new features represents a two-sided coin for MEV searchers. Indeed, some spamming bots would be forced to shut down since the local fee market will make it unprofitable to massively spam txs. However, improving the network means more and more users are attracted to use it. This has the immediate consequence of also increasing the total amount of MEV, allowing the chain of implications to continue by incentivizing competition around MEV and “inviting” new searchers to step in.
One of the main problems that can worsen an AMM’s functionality is pool congestion. This is because if there are too many txs happening on a specific pool, users may experience a worse trade due to pool unbalancing. This is why arbitraging is a sort of service that normalizes DEXs functionality. But, despite the fact that we know MEV is happening on Solana, where are the greatest opportunities? In other words, what are the DEXs with the highest pool congestion, and who is “solving” it? To answer these questions, we built an MEV dashboard on Dune Analytics. This is because, by looking at the exchanged volume, — using Solscan — you can definitely have an idea of where the congestion is, but nothing is clear when the question is if searchers are solving for it.
Our preliminary research shows that in 10 days (from July 16th to July 26th), the paths with the highest extracted MEV on Solana were live on Orca and Raydium with a lower bound of 20,775 USD extracted, see Fig. 2.5. There were 68 MEV extractors on these cross DEXs during the analyzed period, thus not a great number in terms of competition. Fig. 2.6 shows how the extracted revenues are concentrated among a few searchers. Precisely, 5 different accounts extracted 80.1% of the total MEV.
It is worth mentioning that none of the studied DEX combinations show a uniform distribution in terms of MEV opportunities, according to what we show in Fig. 2.2.
If we extend the analysis by looking at the USDC Token Accounts belonging to the most profitable MEV searchers, we have that 7 accounts were able to extract 95.6% of total extracted MEV, see Fig. 2.7. Two of them, GjT…m2P and G9D…y2m, interact with the same smart contract, which may indicate that these two accounts belong to the same user. Since these accounts are in the top 7 accounts, this means that it is likely that only 6 users were able to extract 95.6% of the total extracted MEV.
By deep diving, we also found two accounts interacting with a smart contract with clear reference to Jito, Jito…HoMA, with a total extracted MEV in 10 days of 3,342.30 USDC (at time of writing), over a total of 158,132 USDC extracted — i.e. 2.1% of the total amount.
We already stated that, on PoS networks, MEV can be seen as a business model since validators can share a portion of the extracted amount with their delegators. However, as shown in Sec. 2.1, this sometimes can constitute a deal that does not truly mean high returns. MEV revenues are strongly correlated with market conditions and DEXs’ usage, meaning that we’re unable to estimate a fixed income to share with delegators. Further, if competition does not grow fast, the promise of sharing revenue with delegators may bring a centralization problem.
To assess this statement, let’s try to formulate a “gedanken-experiment”. Imagine that the volume exchanged by DEXs on Solana grows by a factor 30, and assume that there is only one validator extracting MEV and redistributing the revenues to delegators. The implication of the increased volume is that MEV also increases. Indeed, a factor of 30 means that in 30 days the DEX’s volume on Solana is greater than $30B, and assuming that the 0.04% of it is MEV — as it happens on Ethereum — this means more than $144M yearly. The implication of having only one validator playing this game is that the extracted amount also increases, making the delegation to them an appealing deal. We can just think that a validator with ~2% of the total stake can extract an MEV of ~ $2.9M yearly. Once the delegation starts to concentrate around a single validator — the sole player — again we have a boost in MEV revenues, since the leader schedule is “stake-dependent” on Solana. This is because the revenue per block is not uniformly distributed, so a higher stake means an increased likelihood of capturing a rare juicy opportunity, pushing up the median of the extracted MEV. If there is no competition, this gedanken-experiment has a single outcome: concentration of stake — i.e. centralization.
Risks become higher if one considers that at the moment Solana is one of the fastest blockchains and that future development aims to improve this even further. The high number of processed txs per second could pave the way for prop firms to enter the market, meaning that more SOL can be delegated to a single validator — the winner of the MEV war.
This, without any doubt, points toward the necessity of building competitive validators for what regards MEV extraction. Once Jito delivers its third-party client for Solana that’s been optimized for efficient MEV extraction (plus its bundle), the risk of centralization can be mitigated. However, even with decentralized block building, as Flashbots aims to achieve with MEV-boost, we remain still far from a definitive solution. Indeed, such an environment makes it easier for builders to buy the blockspace of all validators and thereby isolate the centralization to the builder layer, see e.g. here. At the moment a decentralized MEV from top to bottom is a chimera. The first step toward this direction would require open-sourcing the MEV-extracting validator, starting collaboration between many validators, in the true spirit of open source. Indeed, it is worth noting that adopting validator products developed — and belonging — to a single entity reduces the problem of stake concentration, but can decrease the network’s censorship resistance. If block production is centralized to a single entity, that may represent an enormous censorship risk, regardless of how many validators participate.
For example, let’s assume that this entity gets adopted by 50% of the stake. Suppose now that this entity is regulated by a specific government, which demands that all transactions are blocked. Then, at best, users would need to get their transactions into the other blocks, but in the worst case, this entity can refuse to include vote transactions that vote on blocks that contain sanctioned transactions. This is a simple example that shows how some MEV strategy outcomes could pave the way for censorship risks.
Before concluding, it is worth mentioning that other possibilities do exist. One of them is to frame MEV-extraction as a service, where it is the protocol itself that takes the MEV and shares the corresponding revenue with protocol-token stakers, see e.g. recent rumors on Osmosis development. Despite this “method” seeming to be less prone to a centralization risk, it remains unclear if the time needed to extract MEV is enough to guarantee the AMM functionality — remember that poor competition means some opportunities may remain there for a “long” time. The outcome is the difficulty of assessing all the details of how this will affect the future of the chain.
This article aims to collect some thoughts on how framing MEV may affect the future of PoS ecosystems, focussing on some of its “bad” consequences. Despite the fast development around this huge and complex topic, we at Chorus One are continuously researching this topic with an eye to the future: the healthiness of all networks is always our first priority.
If you’re interested in framing the topic and require research/advisory services on MEV, you can contact our Research Team at research@chorus.one
Avalanche has a thriving, friendly, and engaging community. On top of that, it also has the quickest and most valuable bridge solution to and from Ethereum, with BTC onboarding shortly. Avalanche is fortunate to have a team that consistently produces and executes at the top level. It’s great for validators like us too. There’s no slashing and rewards are dependent only on uptime. Currently, the annual staking rewards are at 9.1%. This makes locking AVAX to stake appealing. The thriving ecosystem is already on display, with liquid-staking now accessible via BenQi (sAVAX, $179M in TVL) and two additional solutions on the way: LAVA and Eden Network + YieldYak. Lido is also building its liquid staking implementation for AVAX. A competitive DeFi landscape is also in operation, including TraderJoe (DEX, $179M in TVL), Platypus (stable swap, $155M in TVL), Aave (lending, $4.64Bn in TVL), and many more. Subnets now allow innovative technologies in both consensus and horizontal scalability architecture to join the network. To make the experience complete they even provide VMs as free open source code ready to be picked up by companies wishing to join the subnet movement.
Avalanche mainnet is made up of two blockchains (C-Chain and P-Chain) and one DAG (X-Chain for ultra-high TPS). These are two types of distributed ledger technologies (DLTs). The P-Chain is responsible not only for dealing with Subnet and all validator information but also to create new subnets and blockchains.
Although the term “subnet” is used interchangeably and synonymously with blockchains, subnets are a bit more complex than that. The technical definition of a subnet is as follows:
A Subnet is a dynamic set of validators working together to achieve consensus on the state of a set of blockchains, according to Avalanche’s FAQ page.
Subnets allow anybody to quickly establish permissioned or permissionless networks with unique implementations that are powerful, dependable, and secure. Developers can use AvalancheGo or AvalancheJS, and Ethereum developers can seamlessly use Solidity to launch dApps as it is fully compatible. Avalanche includes features not seen on other chains, such as the ability to choose which validators secure their Subnet activity, which token is utilized for gas costs, bespoke economic models, and more. Subnets, crucially, stay naturally linked with the larger Avalanche ecosystem, do not compete for network resources with other projects, and are accessible in an infinite supply. With standard rules underlying all apps on a smart contract network, Web3 applications may distinguish on user experience like never before. A similar approach can be found in Cosmos with Saga and their “chainlets” approach and in Ethereum with Skale.
GameFi, a common phrase in the crypto-verse, is a combination of the words “Gaming” and “Finance.” It covers the gamification of the working system in order to generate profit via play-to-earn crypto games. In GameFi games, items are represented by NFTs. Users may boost their earning potential by levelling up and upgrading their characters, as well as participating in tournaments. As an example, players in Axie Infinity (arguably the biggest GameFi game in 2021) earned more than $1000 worth of $SPL a month before it suffered a hack. Many of these blockchain games are communities where players may earn tokens to swap for money. It’s remarkable to watch blockchain games with a few hundred players in 2013 turn into top-grossing games like Axie Infinity with hundreds of thousands of dollars in daily trade volume. And this is just the first generation of games on blockchains.
Adoption has skyrocketed over the past years. With a large number of retail investors as well as big companies like Microsoft, Nike, Meta and many more already involved, the metaverse market is expected to grow significantly. Major investors such as Gala Games and C2 Ventures formed a $100 million venture fund for GameFi. Solana Ventures and others also launched a $150 million fund by the end of 2021. More recently, Framework Ventures has allocated half of the $400M fund to Web3 gaming. As evidence of the blockchain gaming industry’s expansion, the blockchain games and infrastructure business attracted over $4 billion in venture capital financing in 2021 alone. Blockchain gaming has grown by 2,000 percent in a year, according to the conclusions of a joint report by DappRadar and the Blockchain Game Alliance (BGA). Although this was prior to the latest crypto meltdown. The scenario might be extremely different right now. However, the crypto gaming business has already received $2.5 billion in investment this year; if this trend continues, it might reach $10 billion by the end of 2022. The report also states that blockchain games drew $1.22 million in unique active wallets (UAW) in March, representing 52% of industry activity. With all of the various technologies collaborating to build a self-sustaining ecosystem, the blockchain gaming sector is poised to become a significant income source and probably the first real utility for blockchains outside payments.
The key advantage of using AVAX for GameFi is the three-pronged structure, which comprises validators and subnets using the P-Chain. Subnets let projects create their own application-specific blockchains (ASBs) that do not disrupt the rest of the chain. As a result, no single game utilizes the whole network bandwidth. GameFi on Avalanche offers the best chance for blockchain games to thrive in their intended setting. Avalanche is also great for creating NFTs, which makes digital assets like NFTs easily available for P2E games or the metaverse. Users can utilize Avalanche to establish their own localized chains that run independently of other chains, allowing them to sandbox their own knowledge and technology for the benefit of their own efforts. Most developers use their own token for gas on their subnet, however, a subsidised gas fee is also an option. Avalanche allows network developers to utilize whatever virtual machine they want or to create their own. You may use EVM or any other VM you like. Aside from the EVM and AvalancheVM, Avalanche now provides SpacesVM (key/value storage), BlobVM (binary storage), TimestampVM (a minimum viable VM), and others are in the works. Modularity rules the roost. Observing web2 games moving into web3 through subnets is a great place to start.
It is worth noting that Avalanche gaming developers are taking a Play-and-Earn method rather than a Play-to-Earn approach. This emphasizes the necessity for the game is enjoyable and long-lasting.
Overall, blockchain games continue to be one of the most appealing parts of the dApp market. Although demand for blockchain games looks to have peaked, gaming dApps continue to drive most of the industry’s on-chain activities. Notably, subnet games like Crabada and Defi Kingdoms are still drawing players even in a difficult 2022.
VCs and investors are pouring money into Web3 gaming ventures at an all-time high pace. Furthermore, financial firms like Morgan Stanley have assessed the metaverse’s economic potential to be at least an $8 trillion business. The Sandbox’s second Alpha season, Decentraland’s Fashion Week, and the overwhelming demand for NFT Worlds indicate a positive future for GameFi. However, security risks such as the Ronin bridge vulnerability and the difficulties of attaining full interoperability remind everyone interested that widespread adoption is not yet here. Avalanche Foundation believes that subnets like Shrapnel and TimeShuffle are the solution for the next generation of gaming, thus it launched Avalanche Multiverse last March, a $290 million incentive program to accelerate the growth of the new Internet of Subnets.
Solana has announced three main changes in its mitigation plan to address the stability and resilience of the network:
The measures are targeting the intense traffic responsible for two out of the three recent incidents. Although the changes being proposed by Solana developers are considered abstract or deeply technical for the general part of the community, the concepts are not completely new, being imported from other already mature systems. In this article, we will try to break down the technicalities and explain them in simple terms.
The current Solana client version for validator nodes (v1.10) already paves the way for some of these improvements to be iterated on until optimal market fit. Fee prioritization is targeted for the v1.11 release, according to the official announcement.
Solana used to adopt the User Datagram Protocol (UDP) for transmitting transactions between nodes in the network. Nodes send transactions through UDP directly to the leader — the staked node responsible for proposing the block in that particular slot — without a previous connection being established. UDP does not handle traffic congestion or delivery confirmation for data. In situations of network congestion, the leader is unable to handle the volume of incoming traffic, which means some packets get dropped. Even at quiet times, some level of packet loss is normal. By sending the same transaction multiple times, users have a greater chance that at least one of their attempts will arrive.
In contrast to UDP is the Transmission Control Protocol (TCP). TCP includes more sophisticated features but for this to work, it requires a session (i.e. a known connection was previously established between the client and the server). The receiver acknowledges (“acks”) packets and the sender knows when to stop sending packets in case of intense traffic. TCP allows for re-transmitting lost packets, once the sender stops receiving acks, the interpretation is that something must be lost, so the sender should slow down.
TCP is not ideal for some use cases though. In particular, it sequences all traffic. If one portion of the data is lost, everything after it needs to wait. That is not great for Solana transactions, which are independent.
QUIC is a general-purpose protocol which is used by more than half of all connections from the Chrome web browser to Google’s servers. QUIC is the name of the protocol, not an acronym.
QUIC is an alternative to TCP with similar features: a session, which then enables backpressure to slow the sender down, but it also has a concept of separate streams; so if one transaction gets dropped, it doesn’t need to block the remaining ones.
Solana is a permissionless network. Anyone running a Solana client is a “node” in the network and is able to send messages to the leader. Nodes can operate as validators — when it is signing and sending votes — and (or) they can expose their RPC (Remote Procedure Call) interface to receive messages from applications such as wallets and DEXs, and send those to the leader.
The leader listens on a UDP port and RPCs listen on a TCP port. Given the leader schedule is public, sophisticated players with algorithmic strategies (“bots”) are able to send transactions to the leader directly, bypassing any additional RPC nodes that would only increase latency. With the leader being spammed, the network gets congested and that deteriorates performance. The UDP port used by the leader will be replaced by a QUIC port.
Quality of Service (“QoS”) is the practice of prioritizing certain types of traffic when there is more traffic than the network can handle.
Last January, after Solana faced performance issues as automated trading strategies (aka “liquidator bots”) spammed the network with more than 2 million packets per second, mostly duplicate messages, Anatoly Yakovenko mentioned in a tweet that they would bring the QoS concept to Solana.
The Leader currently tries to process transactions as soon as they arrive. Because IPs are verifiable through QUIC, validators will be able to prioritize and limit the traffic for specific connections. Instead of validators and RPCs blasting transactions at the leader as fast as they can, effectively DoS’ing the leader, they would have a persistent QUIC connection. If the network (IP) gets congested, it will be possible to identify and apply policies to large traffic connections, limiting the number of messages the node can send (“throttle”). These policies are known as QoS.
Internally, staked weighted QoS means queuing transactions in different channels depending on the sender, weighted by the amount of SOL staked. Non-staked nodes will then be incentivized to send transactions to staked nodes first, instead of sending directly to the leader, for a better chance of finding execution, since excess messages from non-staked nodes will most likely be dropped by the leader.
According to Anatoly validators will be responsible for shaping their own traffic, and applying policies that will avoid vulnerability. For example, if a particular node sends huge amounts of transactions, even if they are staked, validators can take action, ignoring the connections established with this node in order to protect network performance.
Solana fees are currently fixed and charged for each signature required in a transaction (5000 lamports = 0.000005 SOL). If there is high competition in a specific market, users face the risk of not getting transactions executed. With a fixed transaction fee, there is no way to communicate priority or compete by paying more to get their transaction prioritized. Without alternatives, users (usually bots) spam transactions to the leader (and soon-to-be leaders) in hope that at least one of them is successful. In many situations, this behavior generates more traffic than what the network can process.
A priority fee is soon to be included in Solana, allowing users to specify an arbitrary “additional fee” to be collected upon execution of the transaction and its inclusion in a block. This mechanism would not only help the network to prioritize time-sensitive transactions but also tends to reduce the amount of invalid or duplicated messages sent by algorithms since speculative operations can become unprofitable with an increase in the total cost.
The ratio of this fee to the requested compute units (the computational cost to the program to perform all operations) will serve as a transaction’s execution priority weight. This ratio will be used by nodes to prioritize the transactions they send to the leader. Additional fees will be treated identically to the base fee today: 50% of the fees paid will be collected by the leader and 50% will be burned.
At this point, you could think of several blocks being filled only with transactions targeting an NFT mint. However, there is a limit time for each account to be locked for writing on a single slot (600 to 800 milliseconds). The remnant block space can be filled with transactions writing in different accounts, even if they offer a smaller priority fee. High-priority transactions trying to write to an account that has already reached its limit will be included in the next block.
Each Solana transaction specifies the writable accounts — the portion of the state that will be modified. This allows transactions to be executed in parallel, as long as transactions are independent, i.e. do not access the same accounts. If two transactions write or read to the same account, these two transactions can not be processed in parallel, because they affect the same state.
The Solana team argues that the priority fee will then behave as parallel auctions, affecting only the “hot market” instead of the global price, allowing the fee to grow for a specific queue of transactions trying to write in that account only.
How does the user know the fee to adopt to get a mint? RPCs nodes will need to estimate an adequate fee, most likely using a simple statistical method, for example averaging the actual cost of similar transactions in previous N blocks, or even a quantile. The optimal method will depend on the market, and whether fees for similar transactions are more volatile (high demand) or stable (less demand).
In practice, the priority fee can have a global effect, if the parallel auctions are not implemented on the validator client. With RPCs and users being responsible for arbitrarily setting it, during high intense traffic, applications will likely try to get priority even though they do not interact with the “hot market”, causing an increase in the fee price for other lower demand dApps.
Fee prioritization is targeted for the v1.11 release, according to the official announcement.
The present article covered the three pieces Solana is actively working on to deal with congestion issues, which include changing the communication protocol from UDP to QUIC, adding stake-weighted QoS for transaction prioritization and a fee market that increases fees with high demand. All of these 3 improvements aspire to improve the performance of Solana, which has been experiencing degraded performance quite often.
We hope it was possible to clarify these concepts and understand the motivations and choices being made. Exploring Solana source code would be an essential next step to investigate the exact metrics being implemented in QoS to select or drop transactions or the mechanism behind the increase (and decrease) of fees and other questions that remain unanswered.
I would like to thank the Chorus One team for the enlightening discussions and knowledge sharing, especially Ruud van Asseldonk for the technical review, and Xavier Meegan for the support.
.png)
