February 12, 2024 - 7 min read
Unleashing the potential of Web3 means pushing the limits of what blockchains do and scaling infrastructure layers to meet the demands of real-world use cases.
In the age of DeFi, things move rather quickly. Fast transaction leads to a smoother and more responsive user experience for DeFi participants. Users can interact with DeFi protocols seamlessly, without frustrating experiences like degraded network performance due to congestion or worse, transactions not going through at all. For dApps to settle transactions with fast finality isn’t just an advantage—it’s becoming essential.
The phrase time to finality denotes how long it takes from the moment the round leader proposes a block to the moment that same block is committed to the blockchain’s immutable ledger. This translates to the network being highly secure, highly responsive, always available, and confirms trades in the blink of an eye.
Consequently, quickly and securely achieving finality have been some of the prominent focal points for the research behind Supra’s vertically-integrated stack of infrastructure. To do so, building the ecosystem to securely minimize latency helps secure the best environment for dApps to deploy and scale their operations.
Of course, scalability, network responsiveness, and fast confirmations form the foundation needed for real-world use cases to become practical, and for mainstream adoption to take hold. Let’s dive into a few of these concepts and elucidate how Supra’s novel consensus mechanism Moonshot sets a new gold standard for distributed networks when it comes to block finalization throughput and block commit latency.
Essentially, it is the delay between user action and the network’s response to the action. In DeFi, a token swap is submitted to a blockchain’s mempool and needs to be executed rather quickly, not only for the user experience itself but in order to execute the trade at the correct price. For a simple swap, a small delay is often tolerated, yet seconds or even milliseconds can make all the difference if the price is in flux.
That is, swapping a token whose price is in the midst of a volatile price swing could mean a transaction is executed at an unfavorable price, called slippage. It could also fail to execute altogether if the transaction isn’t settled quickly enough. Latency which results in meaningful slippage can be devastating in terms of missed opportunities, and is simply unacceptable given the advancements made throughout Web3 in recent years. Experiencing latency issues while interacting with a blockchain network can lead to various problems, which we’ll go over briefly below.
First, latency delays the settlement of transactions on any given network, but blockchains especially. This delay is frustrating for users, especially in time-sensitive situations like making payments at a point of sale or swapping tokens while prices are volatile. It’s simply unacceptable in the modern era, and given that Web3 aims to upgrade legacy financial services.
For example, let’s say a user submits a smart contract agreement to swap a certain amount of one token for another at a given price. If there is too much latency, executing that transaction would take longer than expected. By the time your transaction is available to be included in the next block, the price could have shifted, resulting in a partial execution or a failure to execute the swap at all.
In addition, latency always results in a poor user experience as they face not only experience delays, but in other cases the network times out altogether. This can be a disaster when trying to interact with a blockchain-based app, say during the initial sale of a popular NFT collection on Ethereum or Solana.
In many cases, entire blockchains had to be rebooted, so to speak, and may have ended up deterring users from adopting blockchain sooner without having the bad experience, arguably doing more harm than good. Users want to know that their transactions went through, and they would surely lose their patience quickly when it comes to lost funds.
It is easy to sympathize with this position, as veteran crypto users know the feeling of dread after transferring funds and not seeing them show up immediately in the intended wallet. Reducing latency means reducing the stress experienced by digital asset users and issuers.
Worst of all, latency introduces extra security vulnerabilities, particularly in distributed networks like blockchains. Latency hinders the ability of nodes to reach consensus quickly, increasing the consensus-related issues which might fork the network or prevent the commitment and finalization of new blocks.
Even relatively small delays create opportunities for malicious actors to exploit gaps in security protocols, potentially leading to issues such as double-spending or internal collusion amongst nodes. In other words, if the delay allows for some participants to gain asymmetrical knowledge of forthcoming on-chain events before they occur, those participants could potentially game the system accordingly to gain unfair advantages.
Fortunately, all of Supra’s services benefit from the groundbreaking research and efforts to launch our Chained Moonshot Consensus protocol. Moonshot is able to produce blocks at network speed, and achieves deterministic finality in only 3 messages between nodes on a per-block basis when on a happy path.
In all tested configurations, Moonshot beats the state-of-the-art Jolteon (the core consensus algorithm behind Aptos) in several key metrics, boasting roughly 50% higher throughput and 54% lower latency, on average. This includes both happy path scenarios as well as fallback path evaluations.
In fact, Moonshot is the first consensus protocol producing blocks for every network hop. In theory, this is the maximum possible block throughput for randomized leader-rotating BFT algorithms, since safely doing so with deterministic finality has heretofore not been considered possible.
At this point, Moonshot is literally pushing the physical limits for leader-based BFT consensus protocols that tolerate partially synchronous networks. To put it short, these are ideally suited for use in global, decentralized networks. Thus, Moonshot is currently unbeatable in block finalization throughput and block commit latency. This step-function improvement is a serious advantage that Supra, as a vertically integrated L1, has over any other project of its kind.
What’s more, for the very first time a consensus protocol of such complexity and performance as Supra’s Moonshot has been formally verified to guarantee safety using Microsoft’s Ivy Verifier. This is the gold standard in terms of protocol security. Through Ivy’s Formal Verification, Moonshot Consensus has been mathematically modeled and reasoned to guarantee NOT to fork even when run for an infinite duration for all sizes of the network, as long as the Byzantine or malicious validators are less than 1/3rd of the network.
First off, Supra’s Commit Moonshot is a rotating-leader-based consensus protocol, meaning that every validator has an equal likelihood of being elected to propose the next block. Upon being elected, the leader sends its block proposal to all of the validators for voting.
Next, validators verify the block and cast their votes to form a consensus so long as there are no signs of abnormal behavior. Validators that observe votes from 2/3 + 1 of their peers consider the related block to be certified and send a second vote for it called a Commit vote. A validator that observes Commit votes from 2/3 + 1 of its peers commits the related block.
This process is sufficient to ensure that all honest validators always commit the same blocks; an essential security property—however, it is not unique to Moonshot. Instead, Moonshot is unique in that it allows the next leader to propose its own block as soon as it votes for the proposal of its predecessor. Consequently, efficiency is increased by a large margin, making Moonshot a superior and more scalable design compared to its peers.
This differs from previous protocols, which require the leader to wait at least until its predecessor’s block has been validated before proposing a subsequent one. This allows Moonshot to make progress towards consensus on multiple blocks simultaneously, increasing its block throughput significantly and decreasing transaction commit latency correspondingly.
Supra is the first project in the industry to have a credible and elegant roadmap to vertically integrate high-throughput, low-latency consensus with native Oracle, dVRF, Bridge, and Automation Services. As such, devs and end users will get to leverage novel and state-of-the-art DeFi primitives that are only possible with our infrastructure stack.
Supra’s research teams, led by Dr. Aniket Kate, are experts in cryptography and world-renowned as DKG (distributed key generation) pioneers. Our Class-Group DKG primitive outperforms the state-of-the-art improving all threshold systems, especially in blockchain settings. Consequently, Supra is advancing knowledge in the field both in academic literature and practical applications.
Supra is more than just powered by Moonshot Consensus. Supra is also powered by and composed of complementing protocols, all homegrown and novel. From DORA the Oracle service to HyperNova and HyperLoop cross-chain communication protocols to and decentralized Verifiable Random Function services, and even automation services, they all fit elegantly together into a cohesive stack. Having said that, many services are modular as well.
None of these modular yet vertically-integrated services are forks, and they all work together to create synergistic, emergent value. Together, they can be effectively wielded by developers to make their dApps faster, stronger, and more effective overall. Value creation begets more value creation, and end users benefit from a superior experience when it comes to network security, availability, and responsiveness.
Sign up for the Supra newsletter for company news, industry insights, and more. You’ll also be the first to know when we come out of stealth mode.