New concept "super modularization": How is it different from classic modular blockchain?
Celestia vs. KIRA?Author: ChainFeeds
On December 12, 2023, Celestia announced the integration of its data availability layer with the Chain Development Kit (CDK) from Polygon Labs, reigniting community discussions about modular blockchains. Celestia defines "modular blockchain" as a blockchain that outsources at least one component of execution, consensus, or data availability to an external independent chain. In addition to the classic concept of modular blockchains, the Cosmos ecosystem L1 project KIRA has proposed a brand new concept—hypermodularity.
New Concept: Hypermodularity
Hypermodularity not only achieves modularity at the software level but also emphasizes the separation between modules at the network level. The software layer focuses on the functions and logic within the system, while the network layer focuses on communication and connections between system nodes. Many modular chains share a limited number of operators, and the homogeneity among operators can lead to high coupling, making the system vulnerable to attacks and increasing maintenance difficulty. Hypermodular design allows individual components to operate in independent modules, where different contract modules can adopt different security mechanisms and protocols.
KIRA is an L1 built on Tendermint and Cosmos-SDK, adopting a "hypermodular" architecture. Its investors include TRGC, NGC Ventures, Math Wallet, among others. Its advisor is Alessio Treglia, the Engineering Director from Tendermint. KIRA provides developers and users with a more efficient and flexible choice through an architecture and consensus mechanism that differs from current modular blockchains.
MBPoS Consensus Mechanism
KIRA has introduced the Multi-Bonded Proof of Stake (MBPoS) consensus mechanism. Traditional staking mechanisms typically only allow a single native token. MBPoS allows staking of multiple assets, including NFTs. By permitting the staking of various assets, MBPoS facilitates larger capital inflows and provides a more flexible, secure, and incentivized consensus mechanism. When one asset faces risks or market fluctuations, other staked assets can still maintain the network's stable operation.
Users participate in network security by staking assets and earn income from two sources: block rewards and a portion of transaction fees. A cap on earnings is set to ensure network stability and prevent certain participants from controlling the entire network through excessive token issuance. Additionally, KIRA issues derivatives of staked tokens, known as Staking Derivatives, or the well-known LSD, which ensures that all staked tokens have liquidity, tradability, and transferability.
Different Architecture
Celestia was the first to propose the concept of modular blockchains, decoupling blockchains into three layers: data, consensus, and execution, where all three layers in a monolithic blockchain are handled by a single network. Celestia focuses on the data and consensus layers, while L2 can allow Celestia to handle the data availability layer (DA) to reduce interaction gas fees. For example, Manta Pacific has adopted Celestia as its data availability layer, and according to official news from Manta Pacific, the DA fees decreased by 99.81% after migrating from Ethereum to Celestia.
Monolithic Blockchain vs Modular Blockchain
Celestia uses light nodes to access data, but light nodes need to communicate frequently with full nodes to retrieve data. While this can reduce node resource requirements, communication between nodes in a large-scale network may suffer from delays, affecting efficiency.
KIRA has designed what they believe to be a unique layered structure. This structure involves each modular sub-component (such as DA, execution, etc.) being operated by the same group of validators/nodes. The same crowd participates in validating the security of the entire blockchain while also choosing to engage in the validation and execution of specific applications. In this setup, the system can precisely identify which nodes are running specific applications and verify whether those applications are executed correctly. By clearly knowing who should own the data and who does not need it, the system can maintain state at the maximum replication level while avoiding unnecessary duplication, thus improving efficiency.
KIRA Architecture Diagram
KIRA is divided into user layer, execution layer, and validation layer. These three layers communicate through a content access layer. The content access layer is the cornerstone of KIRA's design, acting as a middleware system between client-hosted front-end applications (static IPFS pages) and the back-end (serving as the settlement layer blockchain).
This middleware system, named INTERX by KIRA, is a decentralized API that allows dApps, interactions, and data state access without relying on any third-party services (such as light clients). INTERX also allows applications to use protocols like TCP (for reliable, ordered data transmission) and UDP (fast but unreliable communication) for output, providing applications with more flexible communication methods, which is especially important for highly interactive applications (such as games).
INTERX has two modes: execution mode and fishermen mode. In execution mode, INTERX acts as an API proxy between the user layer and execution layer, forwarding data changes (transactions) generated by dApps to a single validator (leader) for execution. In fishermen mode, INTERX acts as a broadcaster between the execution layer and validation layer, broadcasting data changes generated by dApps to multiple validators (fishermen). If fishermen observe improper behavior or errors from the executor, they have the right to initiate a challenge. If the fishermen's challenge is proven correct, they will receive a reward. Conversely, if their challenge is incorrect, they may face penalties, including losing part of their stake.
In addition to INTERX, KIRA has two products
MIRO: The front-end application and network wallet of the KIRA network, allowing users to interact with the KIRA blockchain through the decentralized API INTERX. It provides an easy-to-use interface for managing KIRA accounts and assets, making it suitable for a wide range of users. When the MIRO web application page loads in the browser, all operations executed on the local computer occur locally. There is no need for any hosted server or access to the internet beyond any local or public INTERX node's IP.
SEKAI: Responsible for handling the interaction logic of all on-chain applications in KIRA, such as processing transactions and state transitions, executed by consensus nodes (validators). Validators monitor the operations executed by the executor. If improper behavior or errors are observed, they have the right to initiate a challenge.
Related Technologies
In addition to the consensus mechanism and architectural design, KIRA also introduces several new concepts at the technical level, each of which involves more technology and requires the team to disclose more details in the future:
Virtual Finality Gadget (VFG): A mechanism used in applications to verify transaction finality, providing developers with the ability to customize verification logic, allowing different validator nodes to use different, non-public verification strategies to validate transactions. This diversity and privacy in verification strategies make the system more robust and harder to exploit by malicious actors, as they cannot predict all verification strategies.
Pessimistic Rolldowns: A type of Rollup that, with the help of VFG, can execute off-chain, deterministic, and non-specific language code, while achieving faster finality and settlement times than zk Rollup and Optimistic Rollup.
Cross-Application Messaging (XAM): Facilitates communication between different Rollups. Through XAM, decentralized validators, governance DAOs, and direct token minting on L1 can be created, seamlessly integrating functionalities provided by other applications.
Metafinality: The core concept of Metafinality is to establish consistency with external networks and systems across multiple blockchains through some mechanism or protocol, simplifying cross-chain and cross-application integration. Users or systems can more easily obtain a unified view of the entire multi-chain system without needing to run nodes on each chain.
Challenges
In terms of economic model design for blockchain systems, modular blockchains like Celestia have simplified designs compared to traditional monolithic blockchains, such as Ethereum. However, this modular design has not been sufficiently time-tested, so its long-term economic benefits remain to be observed. Ethereum's gas mechanism covers the consumption of block space and computational resources, while Celestia itself does not include a computation layer, allowing the pricing of computational costs to be borne by downstream protocols. On the other hand, KIRA, representing the hypermodular concept, requires more time to validate the effectiveness and rationality of its tokenomics design.
Moreover, from the perspective of end applications, Celestia's downstream applications and products aimed at end users will require more time for construction and development. As an emerging platform, KIRA needs more time to build its ecosystem and applications. The demand and development of these applications are key to capturing real value and reflecting the value of the KIRA protocol layer. Only when application demands are realized can the value of KIRA's protocol layer be truly transformed.
Mainnet TBD
In July 2023, KIRA launched the testnet Chaos Network. Unlike traditional test networks, ChaosNet features account balances that remain consistent or do not change significantly between new iterations. In the initial phase, the core team operates ChaosNet, and over time, more autonomy will gradually be granted to the community, including proposing upgrades, organizing governance, suggesting changes, and electing new validators and governance members.
Regarding the launch of the mainnet, although the core components of KIRA have been completed, the launch costs and coordination efforts will be very high without sufficient infrastructure support. The KIRA team has stated that they are actively working to reduce the costs of launching applications. Additionally, with KIRA's current market value (28 million), launching the mainnet is not economically feasible. Founder Asmodat has indicated that the launch time of the mainnet will be influenced by the attention and demand for the KIRA project.
