Disrupting the Blockchain Industry: Behind the Making of KYVE’s Trustless Data Solution
With KYVE being a one-of-a-kind Layer 1 blockchain that provides innovative, decentralized data tooling to the public, building it required an entirely new thought process for data in Web3.
After gaining support from the top L1s off of just the idea, then over 100k lines of code written across three different repositories, extensive testing by over 43,000 distinct users, and customization of certain Cosmos SDK features, what began as a simple bounty script has evolved into a complete network aimed at revolutionizing the data industry for the better.
And now, with KYVE mainnet live, we’d like to shed light on our learnings and unique development journey these past two years in getting KYVE to where it’s at today.
From Bounty Script to L1
It all started when KYVE’s Co-Founders John Letey and Fabian Riewe came together to work on a bounty project bridging Polkadot data to Arweave. After creating a successful solution, they realized that with further decentralization and development, the solution could be applied to all blockchains in Web3, aiding the overall data space, creating: KYVE.
They first developed a smart contract on Arweave and launched a testnet. However, it soon became apparent that the amount of data KYVE was archiving at once was almost too overwhelming for the system. So, our team worked on optimizing the way data was archived, allowing us to handle larger amounts of data at once.
After this initial success and gaining support from major blockchain ecosystems such as Avalanche, Solana, Near, and Cosmos, we decided to switch to Ethereum, which led us to Moonbeam. During this time, our team also grew, adding two new developers. Together, KYVE’s Development Team continued to improve the protocol, adding new integrations and adjusting to changes from Arweave to Ethereum. Despite optimizing the protocol to reduce gas fees, block space became a continuous issue, leading us to explore the concept of app chains.
After a lot of evaluation, we found that the Cosmos SDK was the perfect fit for our ideology, and we were able to port over the smart contract logic to Cosmos within just 5 days, with the first official bundle being produced after just two weeks. Our team was very pleased with the result, as we now had a Cosmos-based L1 blockchain offering us the scalability, security, and flexibility KYVE needed.
From there, our Development Team has not only created a strong base consensus layer and unique protocol layer but also has made significant contributions to the blockchain community. What all went into building the consensus and protocol layer?
Building the Consensus Layer
The consensus layer is the backbone of KYVE. Being a sovereign Delegated Proof of Stake blockchain built with the Cosmos SDK and using the CometBFT consensus engine, its main responsibilities include reaching consensus and securing the KYVE protocol layer above it.
To adapt the Cosmos SDK to KYVE, a crucial customization involved modifying the governance to accept both our validator sets. Additionally, we’ve done a lot of modifications to avoid spam, namely on-chain tx fees as well as a min initial deposit into a proposal when submitting. (This is the standard that we see now on the Cosmos Hub, Juno, and Osmosis).
Finally, when building, we always stayed, and aim to continue to stay, on the latest developments of the ecosystem. This made us one of the first chains to run Cosmos SDK v0.46. Doing so aligns with our values in giving back to the community and bringing value to it, utilizing the newly introduced features, and providing feedback when necessary.
All of this came together to create a strong and secure base for KYVE’s protocol layer, the true data lake technology.
Bringing the Protocol Layer To Life
KYVE’s protocol layer sits on top of the consensus layer and is what brings to life KYVE’s true objectives. The protocol layer consists of data pools specific to types of data like Cosmos Hub blockchain data, and protocol validators that alternate taking on the role of uploading and validating said data.
Main Customizations and Tweaks
Being a one-of-a-kind solution, the protocol layer required a lot of customization and innovation while also ensuring that our core values of decentralization and customizability stayed at the heart of it all. Let’s delve into a few of the main customizations…
When moving to Cosmos, transitioning the raw logic was actually easier than expected. However, with KYVE’s two layers comes two types of validators. Creating an entirely separate set of validators and incentives was definitely the harder part, requiring a special setup and rewriting so that they could interact with the Cosmos chain.
Thankfully, with the Cosmos SDK, it was possible to easily introduce new protocol features like upload timeouts and runtime upgrades to strengthen the layer. Furthermore, our Development Team performed strict unit tests to catch any possible attack vectors and bugs when keeping up to date with the fast-growing Cosmos codebase.
We did run into a few forced changes, such as with certain third-party entities requiring a change to the overall protocol behavior in the case of their service being offline for certain time periods. While adapting to this, our team also reconstructed the inner parts of the protocol to provide maximal security and customization.
KYVE allows for full customizability when providing trustless data; therefore, users can input the necessary information to create the data pools they need. This means introducing unique runtimes that contain business logic tailored to the validation of each specific data set. While these runtimes can be upgraded to maintain customizability, this also means that the validators need to be upgraded as well.
To ease this process and reduce the risk of offline slashing, KYVE has introduced KYSOR. This acts as a supervisor that enables automatic validator upgrades, thus simplifying the experience for validators.
To further enforce maximal security and incentives for protocol validators, the team introduced Inter-Pool Security. This feature decreases the possibility of high-staked nodes taking over a pool by allowing them to be active in numerous pools at once, securing them with their stake and delegation while also increasing incentivization for good behavior via specific slashing terms. In addition, the team implemented the F1-distribution algorithm to efficiently calculate the delegation rewards and slashes.
Protocol Creation Overall
Following our values of decentralization and transparency, our code is fully open-sourced, and the Development Team ensured to bring forward detailed developer docs to share every aspect of the KYVE protocol for its different types of users: token holders, data engineers, Web3 developers, protocol developers, and validators.
“KYVE can be proud of having developed a protocol where data is actually validated between all participants and making it able to store onto decentralized storage platforms, actually being transparent about it throughout the whole process.” Said KYVE Developer, Troy Kessler.
KYVE’s clean modular codebase has also already received acclaim from code reviewers. Not to mention achieving all of this with a team of just three developers (now four with Chris), as opposed to the usual larger team size of around 20 members.
Throughout all this development, the protocol has proved to represent a significant advancement in the Web3 data space, and we’re excited to see how it will continue to grow. It is important to note, however, that our success is not solely thanks to the KYVE team, but also to our amazing network of community members, partners, and investors.
Support From Our Community, Partners, & Investors
The Layer 1 network we first launched was a testnet network called Korellia. This was a large mixing pot of new features and improvements, which required heavy testing, leading to the Korellia incentivized testnet.
Our Incentivized Testnet was a resounding success, providing the Development Team with a valuable opportunity to test the protocol in a simulated, real-world environment while allowing participants to provide direct feedback. The three-month event saw over 43k participants contributing to the network, resulting in over 178k bundles of data (over 82GB of data) created and uploaded by 2,200 different validators.
In addition to the incentivized testnet participants, the Development Team also received valuable support from dedicated beta testers. Both groups played a critical role in providing feedback, running nodes, and identifying bugs and other issues. Without their contributions, it would not have been possible to achieve the level of testing required to develop a robust and reliable protocol, fine-tuning it to ensure that KYVE is as reliable, efficient, and user-friendly as possible.
There’s no doubt that the success of Mission Korellia, coupled with the ongoing support of our community members, partners, investors, genesis validators, and more have been instrumental in KYVE’s development.
The Genesis Validator Program
When preparing for the launch of Kaon, KYVE’s official testnet, it was the perfect opportunity for a dress rehearsal for mainnet launch. This meant banding together a strong genesis validator set and program in advance.
Luckily, from the start, our partners, investors, and other supporters have been following KYVE’s development and looking forward to participating as validators, giving KYVE a great selection of to start with. Not to mention, most of them being some of the best validators in the Cosmos ecosystem, providing our team with great advice and guidance throughout the rehearsal process.
Felix Lutsch, Chief Commercial Officer at Chorus One, stated, “The KYVE stack has the potential to bootstrap many data innovations by enabling developers to outsource collection, storage, and retrieval of data to the trustless KYVE protocol. […] we couldn’t be more excited to support this community with our infrastructure and expertise.”
LOA Labs team team also mentioned,“KYVE Network’s talented team is on a mission to build novel data services on decentralized, trustless blockchain ecosystems. LOA Labs is aligned with KYVE’s vision and supports KYVE’s work in making data a Public Good: accessible and secure for all.”
Final Thoughts
After two years of intense development, we’re thrilled to see where KYVE is at today, and we really look forward to all of the innovation it will bring in the near future. With mainnet now live, our team is just getting started exploring all of KYVE’s capabilities.
From this experience building KYVE, a major takeaway is not to be intimidated by change or the need to learn a new blockchain structure. With a solid understanding of programming languages and IT concepts like networking and encryption, anyone can dive into the Cosmos SDK and start building. At its core, it’s simply a unique data structure waiting to be explored.
About KYVE
KYVE Network is revolutionizing customized access to on- and off-chain data by providing fast and easy tooling for decentralized data validation, immutability, and retrieval. Its data lake archives and validates data in a decentralized way, making trustless data easily accessible via solutions like KYVE’s Data Pipeline.
The network is powered by decentralized uploaders and validators funded by $KYVE tokens and operates as a DAO following our mainnet launch. This past year, KYVE has gained significant support, currently backed by top VCs, including Hypersphere Ventures, Coinbase Ventures, Distributed Global, Mechanism Capital, CMS Holdings, IOSG Ventures, and blockchains such as Arweave, Avalanche, Solana, Interchain, and NEAR.
Interested in the details of each protocol update passed so far on KYVE’s testnet Korellia?
- Protocol update #1
- Protocol update #2
- Protocol update #3
- Protocol update #4
- Protocol update #5
- Protocol update #6
- Protocol update #7
Stay tuned for further updates and developments to Kaon and KYVE mainnet.
