The Interoperability Stack: A Hitchhiker's Guide to Multichain dApps that Don't Suck
We describe the core building blocks of interoperability - General Message Passing, intents, chain signatures, interchain tokens - and how the composition of these blocks delivers a unified Web3.
A joint blog with CJ Cobb and Ben Weinberg.
A special thanks to Curtis Spencer, Maria Shen, and Galen Moore for proofreading and feedback on the post.
The Web 3.0 ecosystem continues to see an explosion of new blockchains launching each year. Horizontal expansion allows the ecosystem to scale, offering cheaper blockspace and a diversity of technical options and communities; the downside is that it fragments liquidity, increases developer maintenance costs across multiple chains and breaks UX. Users need to navigate their journeys through multiple wallets, gas tokens and bridges just to try new applications. Developers end up spending more time deploying and managing their applications across multiple chains. According to the Electric Capital developer report, the number of developers building across multiple blockchains has increased 10x since 2015 and one in six developers writes code for more than three chains.
Today, end-user dApps, such as wallets, prediction markets, or NFT marketplaces are still suffering from a lack of intuitive UX compared with applications on the internet at large. Many of the blockchain industry’s largest dApps are siloed in a single blockchain and require manual work from their users to be able to use them.
Interoperability infrastructure simplifies these experiences – leading the paradigm frequently known as "chain abstraction." Simply put, chain abstraction means developers and users interact with multiple chains as if they interact with one. However, the technical journey towards these unified experiences is not a simple one: it relies on layers of protocols and services working together. In this post, we summarize core components of the multichain Web3 infrastructure stack: the interoperability protocols necessary to achieve chain abstraction. The first principle is to separate the infrastructure and settlement layer from the application-level protocols sitting above it.
The interoperability stack is complex. This blog serves as a systematization of knowledge to help web3 users and developers navigate.
The remainder of the post is organized as follows:
First, in Part I, we describe the core function - General Message Passing - required for interoperability. GMP protocols securely transmit messages and assets across blockchains. They provides base-layer APIs for developers and ensures naming conventions, safety and liveness properties.
Part III serves as the developer’s guide in into specific application-level interoperability protocols. We explain intents, chain-signatures, queries, multi-bridge approvals, token transfers, and interchain governance.
Then, in Part III, we dive into the unified multichain UX achieved via interoperability stacks. We review how wallets, DeFI, NFT, and RWAs can evolve to achieve unified liquidity and seamless UX.
Finally, we summarize with the long-term view of the ecosystem and the end-state user interactions.
Part I: The core function: GMP
At the core of all chain abstraction primitives sit General Message Passing (GMP) protocols that enable secure cross-chain settlement and transmit messages across chains between smart contracts. These protocols define a base-layer API that developers use to send information and assets securely across multiple blockchains. They also implicitly or explicitly define naming conventions like chain names and address formats (see CAIP-10 and EIP-3770). As such, GMP protocols must satisfy robust safety and liveness properties for developers. This is where the base-layer distinctions between various protocols lie: whether they are controlled by a single entity, use multisignature setups, or are open and permissionless. Examples of GMP protocols include Axelar GMP, Chainlink CCIP, Cosmos IBC, LayerZero and Wormhole.
GMP protocols should be minimalistic, to ensure the strongest liveness and safety properties. Leveraging the GMP layer are application-level protocols that simplify cross-chain interactions for developers and users by automating frequently used flows. Examples of these include token transfers, chain signatures, cross-chain governance, multi-approval contracts and intents. We next describe these protocols in detail, and subsequently, in Part III, we review the UX and properties they unlock.
Part II: A deep dive into application-level interoperability protocols
In this part, we review the application-level protocols that make it easier to build application and UX experiences described above. All of they rely and can be instantiated with GMP guaranteeing security and reliable composable cross-chain execution.
Interchain Tokens
Interchain token services are a critical component in the journey toward chain abstraction. Traditionally, tokens were launched on a single chain and relied on bridging protocols to recreate their representation on other chains. These newly created tokens often carried prefixes or suffixes from the bridging protocol, had distinct addresses, and were not easily fungible between different representations. Interchain tokens, however, allow developers to launch a single representation of their token across multiple chains with the same address and name, and provide instant cross-chain support. The token moves and interacts on each blockchain as if it were native there, and supply is managed across multiple such instances. This enables both developers and users to seamlessly integrate the token into one-click experiences, such as depositing a token from chain A into a dApp on chain B with a single transaction.
Examples include Axelar ITS, Layerzero OFT and Wormhole NTT.
Chain signatures
Chain signatures provide a centralized “place” where a user can sign transactions to be accepted by other chains. Typically, to sign for chain X, a user would need to have a wallet and native tokens for gas fees on chain X. However, a system implementing chain signatures abstracts this process. Given a user’s key and a transaction to be executed on chain X, the system signs, propagates and executes the transaction on that chain without requiring the user to obtain the corresponding wallet or explicitly pay gas fees.
What can you do with chain signatures? If a user holds funds across multiple chains (e.g., X, Y, Z) and wants to interact with applications on each chain, they do not need to switch to a wallet on each respective chain. Instead, they can submit a transaction from a central system and have it executed on any of the underlying chains supported by the chain-signatures scheme.
What chain signatures do not solve: Chain signatures, by themselves, do not unify a user’s or an application’s liquidity or compose functionalities across multiple chains. For instance, if a user holds $10 USDC on chain X and $5 USDC on chain Y and wants to deposit $15 USDC into an application on chain Y, that application would need to employ a GMP or basic bridging solution to make that possible. Similarly, an application cannot unify its liquidity and must treat each of its instantiations across chains separately – unless it uses a GMP-enabled service to coordinate pools of liquidity.
How chain signatures work with GMP: Chain signatures can be implemented using GMP interoperability protocols. Any chain that supports GMP can implement chain signatures, allowing any wallet on that chain to act as a "hub" for users to access all dApps across all chains. GMP enables applications to send executable messages to each other across chains. So, if a user requests to use an application on any of the chains (X, Y, Z) supported by the interoperability protocols, a wallet can simply forward that request to the destination chain and execute it.
Limitations without GMP: Chain signatures do not necessarily require GMP and can rely on a single or multiparty protocol to sign transactions across multiple chains. However, without GMP, achieving full composability and liquidity sharing in a decentralized manner is not possible.
Intents
Intents are an application-level feature built on top of GMP protocols, designed to enhance user experience by speeding up token bridging and cross-chain swaps. Typically, an off-chain component – often referred to as a filler, solver, or relayer – optimistically executes the transfer or swap on behalf of a user, providing any necessary liquidity upfront. This process happens very quickly, often within seconds. After the swap or transfer is executed and the user receives their tokens, the filler settles with the protocol by claiming the tokens the user originally locked. This settlement layer is usually implemented via GMP, as the protocol needs to verify which filler fulfilled the order and ensure they can claim the tokens. The settlement step is typically slower because secure cross-chain communication cannot happen instantly. Intents do not replace GMP; they add a layer on top of GMP, abstracting secure message passing away from the user. The user receives their tokens almost instantly, while the filler waits for the GMP process to complete to receive reimbursement and rewards for filling the order.
Squid example: Consider Squid, a token bridging and cross-chain swap platform built on top of Axelar’s GMP protocol. Using the Squid UI, a user can request to swap token X on chain A for token Y on chain B. A transaction is sent to the Squid contract on the source chain, initiating the Axelar GMP flow – a complex process that involves waiting for finality on the source chain, voting on the transaction’s validity on Axelar and creating a proof for the destination chain. This process can take from 30 seconds to several minutes, depending on source-chain settlement times. However, if the user selects the “Boost” function on Squid’s UI, a filler optimistically fulfills the swap/transfer on the destination chain before the on-chain settlement process completes, allowing the user to receive their tokens in seconds. Once settlement is final on both connected chains, the filler is refunded for the tokens provided to facilitate the swap. Read here for a more detailed look at how this works.
Other intent-based bridges: Other intent-based bridges operate similarly, where the user’s request is fulfilled before the GMP process is completed (or even started), and the filler relies on GMP in some form for reimbursement during the settlement phase. For example, Across Protocol uses the UMA oracle to prove that orders were filled by a particular filler, and DLN (deBridge Liquidity Network Protocol) uses the deBridge messaging protocol to send a message from the destination chain back to the source chain to facilitate settlement.
Intents and GMP: Intents and GMP are not mutually exclusive, but rather just different parts of the stack. Any secure and decentralized intent-based bridge must be built on top of message passing in some form. Intent-based bridges are essentially cross-chain applications that use GMP as an infrastructure layer during the settlement phase. These bridges abstract the message passing process away from the user, who gets their order filled quickly, while only the filler needs to be aware of GMP and wait for that process to complete. The application and infrastructure layers are separate and modular though, meaning any intent-based bridge could use any GMP protocol for settlement. For instance, EIP-7683 defines a cross-chain intents interface that can be implemented on top of Axelar GMP.
Multibridge approval contracts
Multibridge approval contracts are built on top of two or more GMP protocols; messages are only considered valid and passed to the destination contract when a quorum of the underlying GMP protocols validates the message. For example, Lido DAO uses both Axelar and Wormhole to bridge wstETH, ensuring that wstETH is only unlocked or minted if both Axelar and Wormhole validate the token transfer. This multibridge approach is flexible and agnostic to the underlying protocols; the set of underlying protocols can be modified (usually via governance vote), and the quorum can be configured to any M of N configuration.
Beyond modularity, the key advantage of a multibridge approach is significantly enhanced security. Even the most secure interoperability protocols carry a non-zero risk of catastrophic failure, such as the software bug that led to the Wormhole hack. However, the likelihood of multiple underlying protocols experiencing such a failure simultaneously is extremely low. The drawback of relying on multibridge approvals comes with higher probability of liveness failures, additional user costs to pay fees for each protocol, and increased maintenance costs. (If a transaction is stuck, multiple protocols might need to be investigated to determine where the failure originated.)
Axelar’s Interchain Amplifier natively supports multibridge approvals within the protocol itself. Message verification within the amplifier protocol is modular, allowing developers to implement verification logic that combines two or more verification methods (e.g., light client verification combined with RPC-based voting verification).
Queries
Cross-chain queries are a modified form of message passing, where instead of a smart contract on chain A sending a message to a smart contract on chain B, the smart contract on chain B queries data from a smart contract on chain A. Ultimately, this still involves sending a message from A to B, with the message containing the query result. The implementation details may vary, but at its core, it relies on message passing.
For example, this could be implemented as a two-way GMP call: a smart contract on chain B encodes a query request as a message and sends it to the smart contract on chain A. The smart contract on chain A computes the query result, encodes the result in a return message, and sends the message back to the original calling smart contract on chain B.
This process can be optimized so that the smart contract on chain B doesn't directly send a message to chain A. Instead, it submits a query request to the interoperability protocol. The protocol interprets the request, queries the contract on chain A, and encodes the result as a message to send back to chain B. The underlying mechanics remain similar to GMP (the query result must be verified, routed, etc.), except that no transaction needs to occur on chain A for the query result to be passed back to chain B.
Interchain governance
Applications that are launched on multiple chains often need to upgrade parameters or their codebase via governance. A native solution would be to redo governance on each chain to update the dApps. However, a more efficient approach is to have governance take place on a single chain, tally the votes, and then apply the updates across all chains. The updates can be communicated via GMP protocols. Uniswap has conducted extensive analysis into various bridging protocols to understand the security and usability implications for its cross-chain governance. Learn how Axelar enabled Uniswap developers to implement one such cross-chain governance solution for its expansion to the Filecoin ecosystem.
Part III: Interoperability use cases and how to improve UX across multichain
Combining the application-level interoperability protocols explained above unlocks a variety of use cases, ranging from unified governance to token creation and wallets. We will review these next.
Wallets
Wallets are the gateway for individual users into the world of Web3. When wallets lack GMP, they deliver a fragmented and limited user experience. The core issues with single-chain wallets are as follows:
Users need to install, manage and subsequently point their wallet to each blockchain network to both see what assets they have on a given chain and to actually use those assets.
On each blockchain they touch, users need to acquire a unique gas token, before they can submit any transactions.
When users want to pool the power of token balances held on various blockchains, they first need to bridge these tokens to the same chain and manually aggregate the balance, before they execute a transaction.
By integrating a wallet with a GMP protocol, the siloed, single-chain wallet can be transformed into a universal wallet, so users can seamlessly interact with dApps and assets on multiple blockchains from a single interface.
To support multiple blockchains with unified UX, wallet developers must connect to a GMP provider such as Axelar as a facilitator of cross-chain transactions. GMP will allow the wallet provider to automate many steps a user would otherwise need to perform manually. These become a sequence of instructions that are executed securely across multiple chains, all from a single user action in the wallet.
In practice, a universal wallet will allow users to:
Aggregate their balances across multiple chains onto a single UI. This will provide users with a much simpler experience to both view their assets and interact with their assets.
Provide users a unique, global, human-readable address where they can receive transactions.
Allow users to interact with dApps directly, regardless of the blockchain underneath. If the dApp is deployed on a chain where the user has no assets to to transact, the wallet can handle multichain asset aggregation in the background.
At some points, users may not even be aware what blockchain they are using – relying instead on their trust in the wallet and dApp developers to select secure infrastructure. This is similar to the way the internet at large works today: most users are not aware which CDN or cloud provider an application has selected. This describes the goal of so-called chain abstraction.
Integrating smart-contract wallets
There are many ways wallets can integrate GMP. One example is applicable for smart-contract wallets. For teams looking to deploy smart contracts as wallets, applications can integrate directly with the Axelar’s contracts and leverage Axelar’s relayer infrastructure to handle seamless transition of the GMP transaction between the source and destination chains. The wallet must import the relevant Axelar contracts to send cross-chain messages and validate inbound cross-chain messages to ensure the incoming message has been confirmed by the Axelar network.
Multichain balance display
When it comes to displaying balances of assets, wallets can make cross-chain queries to the token contract’s instances across multiple chains and render the total balance on a single page. This offers users a much cleaner experience than switching between different chains on a wallet to see their balances on a per-chain basis. With the emergence of Interchain Tokens, users will be able to send tokens and NFTs as-native, right from the wallet interface.
Multichain LP
In the DeFi ecosystem, if a user wants to provide liquidity (LP) for a decentralized exchange (DEX) across multiple chains, they may encounter different token requirements on each chain. A universal wallet can greatly enhance user experience by enabling a single-click experience. This wallet would allow the user to LP from a single chain, while seamlessly handling all cross-chain functionality in the background. As a result, the DEX can access the required liquidity, and the user can LP their assets without navigating the complexities of multiple chains.
Improved gameplay
In blockchain-based games, universal wallets open the door to improved gameplay experience. As games continue to expand to multiple blockchains, scaling a fluid gameplay experience becomes a challenge for developers. In-game assets, which may technically live on separate blockchains, would need to be displayed in one single interface for a more simple gameplay experience. Also, the ability to use those assets across different components of the game that may exist on separate blockchains is crucial for maintaining an integrated and fluid gameplay experience. As with the previous example the universal wallet would abstract away the onus which is currently on the user to need to configure their wallet to have the required assets for each chain they want to transact on.
DeFi
DeFi today has grown to include many different types of applications including DEXs, lending platforms, staking aggregation and much more. However the innovation of DeFi will be limited if major protocols are not able to integrate across different blockchains. In today’s DeFi landscape, key issues are:
Liquidity is scattered across dApps built on various blockchains. Even the same dApp which is deployed across a number of chains does not function as one pool of liquidity. This is particularly an issue for dApps on smaller chains, which can be susceptible to slippage and market manipulation.
dApps provide inconsistent UX on various blockchains. For example, a DEX may allow users to swap between a wide array of tokens on one blockchain but a much more limited set of tokens on another chain. Some of the most popular stablecoins, for example, have no presence beyond just a few blockchains.
Limited composability between DeFi protocols: for instance, a lending protocol on one blockchain may not necessarily be able to integrate with a yield farming protocol on another chain – thus limiting the ability to create novel financial products.
In an ecosystem where a large number of DeFi applications are implementing Axelar GMP, the lines between same-chain and cross-chain transactions will become less distinct. A DeFi protocol will act as a unified application and liquidity pool, encompassing all the different blockchains where it is deployed. This can replace a model where deployments per chain are viewed as separate dApps with different available assets and costs.
One case where this can have a significant impact is in cross-chain staking platforms. The ability to roll out services such as cross-chain staking pools can be a huge opportunity to inject liquidity to help bootstrap new blockchains. For users this offers stakers a simple gateway into earning rewards in tokens that they may have had a difficult time accessing before. For example, users provide funds to Lido contracts on Ethereum which in turn stakes the funds on “exciting new blockchain”, frequent cross-chain messages can be sent between Lido on the two chains to stake new funds for users who are based on Ethereum and pay those users back on Ethereum in a wrapped version of the token that they are providing liquidity for.
Lending platforms could also see significant distribution. One example is pertaining to cross-chain collateral. For lending platforms that live on emerging blockchains finding enough collateral to power these dApps could be an issue. By using cross-chain GMP, dApps can now get access to collateral from a whole new pool of lenders and assets that may not have been available on the more niche chain. From the lender's perspective they now have the ability to earn much higher returns as they can now lend to chains with less liquidity available for higher rewards (and higher risk). On the flipside borrowers on these more niche chains will now have access to way more liquidity than they otherwise would without the GMP integration.
Another example where Axelar GMP can be used is in cross-chain token swaps. Applications like Squid Router are able to offer users a one-click cross-chain bridging experience using Axelar GMP. Where users can easily move a token from one chain into a different token on a different blockchain all in a single click and with one single gas payment.
In the video you can see how Squid provides an intuitive UI for users to select which chains they are transacting between and which token they want to bridge between. The real time saver of using Squid is that users do not need to directly bridge between the same asset on different chains (for example Ethereum-WETH to BNB-WETH). Users can receive a swap like experience where they bridge in one asset on a given chain and are outputted a different asset on the destination chain.
NFTs
Nonfungible tokens (NFTs) have proven to be one of the most popular innovations in Web3. However, without interoperability, NFTs are limited to their origin blockchains and therefore unable to achieve their full potential value and utility.
Value: The owner communities NFT projects have built on a given blockchain will remain isolated on that one blockchain. Take, for example, the popular NFT collectible, Bored Ape Yacht Club (BAYC). This project has built great value for a community of Ethereum holders. As new blockchain communities grow, attracting users away from Ethereum, the BAYC NFT will decline in value if demand is limited to only the Ethereum community, which is going through its yet another existential crisis as of this writing.
Utility: Popular Web3 projects are building their own, dedicated blockchains to improve overall UX. NFT projects are no exception. Pudgy Penguins recently announced their intention to roll out their own blockchain, called Abstract. If Pudgy Penguins holders are now siloed to the Abstract chain which lacks the user base and dApp ecosystem of other popular blockchains, then the ability of the NFT to be used in partnership with other dApps as in-game assets or as collateral for DeFi protocols will be cut short.
NFT + GMP = 😀
Users will no longer need to know (or care) which blockchain their NFT is being used on as NFTs become integrated with dApps across multiple chains. Cross-chain and same-chain transactions will become one and the same for the user.
The NFT will maintain its provenance through both same-chain and cross-chain transactions. If the user who bought a Beeple NFT, originally minted on Ethereum, for $69 million, has sent it to an address on the Avalanche blockchain, users can track the transaction all the way back to the NFTs original home on Ethereum.
The NFT will be able to handle metadata for cross-chain transactions. An NFT’s value often lies in its metadata. The metadata must be transferable along with the NFT.
The NFT will be able to make cross-chain royalty payments. For example, let's go back to the $69 million Beeple NFT that was minted on Ethereum. If down the road the NFT gets sold by a user on Avalanche to a user on BNB, the cross-chain NFT must be able to pay royalties back to the original minter on Ethereum automatically.
Here's a real-world example: the Decentraland NFT Marketplace leverages Axelar for cross-chain NFT purchases. In this brief clip you can see how users can seamlessly purchase an NFT from any chain with one click. Users can now easily tap liquidity they hold on any connected to chain to purchase the NFT. The different networks' gas payments are abstracted away so the user gets that one-click purchasing experience.
How to bridge an NFT
“Bridging” NFTs without native support on each chain is a lot more cumbersome. These are nonfungible assets: unlike dollars or units of ETH, one NFT is not equivalent to another, so "wrapped" representations don't work. (See a guide to cross-chain bridges for how this works with most tokens.) Instead, the NFT needs to be deployed as a native asset on each blockchain where it is to be used or held. To do this, the NFT will need to ensure a consistent and unique token ID across all of these chains. Along with managing the NFT’s token ID, token metadata will also need to be accessed on each chain. Images may also need to be retrieved from on-chain file storage. For example, if the NFT is a Bored Ape, the relevant IPFS pin for the token’s image will also need to be retrieved.
Example use cases for cross-chain NFTs
Cross-chain NFTs open the door to a wide range of new utilities for NFTs. For example:
For dApps using NFT-driven governance, GMP-powered cross-chain NFTs can allow for a larger number of community members able to vote on a governance proposal. Without cross-chain NFTs, the only members who could vote would be the NFT holders on the dApp's original blockchain. This makes participating in governance much more accessible as users do not have to find a way to bridge their NFT to a given chain just for a governance vote. With cross-chain GMP the user can send a cross-chain vote as a GMP message and the dApp’s governance contract can send a cross-chain query to the NFT contract to verify that the vote is coming from a valid NFT holder.
In the collectibles space, NFT Marketplaces could directly integrate with application-level protocols such as multi-bridge approvals – where a transaction is routed through multiple cross-chain verification methods (see below for more detail). In practice, this means the NFT will now be available to users across all chains in what can be thought of as a unified NFT marketplace. This will expand the pool of buyers and potentially increase demand.
Marketplaces where high-value NFTs are bought and sold can allow owners of these NFTs to leverage a multi-bridge approval when moving their NFT between chains. With approval by multiple interoperability providers, the marketplace can ensure the most secure bridging process. For example, the new owner may only want their NFT bridged if both the Axelar and Wormhole networks come to independent, internal consensus of the state of the NFTs transaction. This multi-bridge approval process could be an industry standard for ultra high-value NFTs.
Another example is NFTs as user reputation verifiers. For a user with a strong reputation in one blockchain ecosystem, a cross-chain compatible NFT can provide the ability to enter another blockchain ecosystem with that reputation intact. This removes a limiting factor that could hold back the growth and composability of Web3: established users on certain ecosystems may be hesitant to enter new ecosystems, where they would need to rebuild reputations from scratch. This is especially relevant in contexts such as Web3 games.
Real-World Assets (RWAs) & Stablecoins
A rapidly growing sector of the web3 space, is the tokenization of real-world assets. The massive opportunity has seen projects tokenizing traditionally illiquid assets such as real estate and private credit and bringing it on-chain. By bringing these assets on-chain the assets can be more easily distributed and fractionalized to a large pool of investors.
Although these RWA present a tremendous opportunity, their effectiveness will be limited if they are restricted to a single blockchain.
The amount of liquidity that can actually be injected into these assets will remain limited if the asset is limited to the community of a single chain,
The utility of the assets will be reduced if it cannot access major DeFi protocols across different blockchains. This is an especially large issue for RWA initiatives by larger institutions thare operating on a permissioned blockchain with far less activity than more mainstream blockchains such as Ethereum.
Without the ability to be interoperable, the asset is completely dependent on the blockchain it was tokenized on. This is a particular issue when it comes to RWAs, as these assets tend to be longer term investments that take many years to mature. With an ever changing blockchain landscape, the RWAs could be stuck on a blockchain that may become obsolete going forward.
There are many ways RWAs are being brought into the blockchain ecosystem. Two popular ways to do this are by tokenizing the asset as an NFT or by using fungible tokens. In the case of fungible tokens RWAs are oftentimes used as collateral for stablecoins as in the case of Ondo Finance or Mountain Protocol. Both of these solutions can leverage Axelar GMP in different ways to provide interoperability for their RWA backed assets.
In the case of NFTs, Axelar GMP can be used to mint/burn NFTs between blockchains (i.e., burn on the source chain and send a message to the NFT instance on the destination chain to mint a given token ID there). This unlocks a huge opportunity for NFTs to be integrated with DeFi dApps on different blockchains. With a cross-chain compatible NFT, that asset can now be used as collateral on DeFi apps anywhere.
For stablecoins, the ability to deploy as an Interchain Token offers developers a simple way to deploy their token with built-in bridging functionality from day one. Even in the case where a stablecoin is already deployed, a GMP-based service such as Interchain Token Service allows for a quick integration to provide native, cross-chain functionality for stablecoins. Once integrated, cross-chain stablecoins enhance trading opportunities by enabling stablecoin holders to engage in margin trading and arbitrage opportunities across various blockchains. When combined with protocols using cross-chain intents to facilitate faster transactions, users can capitalize on these opportunities even more effectively due to the faster execution speed.
Summary
In the history of the internet, horizontal scaling within and across networks has been essential for its growth (See “Towards robust distributed systems” and “Harvest, Yield, and Scalable Tolerant Systems.”) No single technology stack could support all possible applications, which led to the development of distributed systems that could scale horizontally across servers and networks. This principle is equally applicable to web3, where relying on a single chain is increasingly impractical given the performance and ecosystem challenges posed by major L1s and L2s.
To scale web3 effectively, horizontal scaling across multiple chains is crucial. While synchronous composability on a single chain is desirable, it is not sufficient to support the diversity and scale of applications that will emerge. Asynchronous composability, enabled by GMP and application-level protocols, provides a robust solution. This approach mirrors the design of the internet, where asynchronous communication and composability have allowed for massive scaling and interoperability across different systems and networks.
Asynchronous composability using GMP ensures that applications can work seamlessly across chains. However, it requires developers to embrace a new paradigm, one that involves thinking beyond the limitations of a single chain and leveraging the tools and protocols that enable cross-chain interactions. This shift allows the web3 ecosystem to scale more efficiently, driving continued innovation in both execution and interoperability layers, in parallel.
Although much work remains to fully realize this vision, the core building blocks are already in place. Developers can begin leveraging these tools to reduce the cost of building and managing dApps across chains, simplifying the user experience while maintaining a simple and unified experience for both developers and users. By adopting these new approaches, the web3 ecosystem can continue to grow and innovate, ensuring it can support the wide range of applications that will define the future of money.