Topology
January, 2025
For the world computer, blockchains excel as decentralized hard drives, great at global persistence but facing inherent scaling limitations in the form of a trilemma. Topology introduces two key components to enhance the world computer: DRP, an Internet protocol that acts as world computer’s RAM by enabling ephemeral real-time peer-to-peer computation, and Mandu, an orchestration blockchain that connects this RAM with decentralized hard drives. These protocols complement the current infrastructure, working in harmony with L1s and L2s alike. Their integration enables developers to build high-performance applications with real-time interaction, end-to-end encryption, and no single point of failure, leapfrogging the blockchain trilemma and advancing the original cypherpunk vision. Through these innovations, Topology reimagines the architecture of the world computer.
The world computer vision
The idea of the world computer is to create a single programmable medium where code and data live and operate correctly forever without centralized intermediaries, enabling trustless and censorship-resistant coordination at global scale.
The World Computer serves as a defensive foundation for humanity's shared interest that transcends geopolitical boundaries. It is designed to resist centralized control and coercion while empowering individual agency. It provides a resilient platform that remains accessible even in the face of powerful actors. Anyone with Internet access can participate, creating opportunities for livelihood, contribution, and community formation that are protected from both corporate and state-level capture.
Public programmable blockchains are hard drives
The world computer vision was originated from Ethereum, which imagined itself as a public decentralized shared hard drive [1]. Indeed, a public programmable blockchain works like a stateless CPU running on a hard drive that allows only sequential writes.
A decentralized hard drive typically exhibits the following properties:
Persistence: Global persistence is guaranteed. Blockchains typically offer the guarantee that any data sent onchain is persisted indefinitely.
Lock-heavy block time: Transactions are total ordered and batched into blocks by global consensus, acting as a single global lock. As a result, transaction latency is dominated by the overhead of this lock.
A single shared transaction log. All transactions related to any part of its state are collectively recorded in a single shared log, typically in the form a merkle tree. To trustlessly recreate the latest state of any particular smart contract requires syncing and verifying the entirety of this log.
Subject to blockchain trilemma: As chains pursue performance improvements, they trend toward centralized block production, where computation occurs in specialized nodes (”thick servers”) operated professionally, while most end users run “thin clients” that access onchain state via REST endpoints.
The primary limitation of these hard drives is to scale in decentralized ways. To achieve higher throughput and lower latency, blockchains would require fewer and more powerful machines to fully participate in the creation and verification of large blocks, which creates a centralization vector.
Faster and modular hard drives are not enough
Multiple base layer blockchains in the recent years have pushed the performance boundary. For example, Solana achieves higher throughput through its proof-of-history mechanism, parallel transaction processing, and higher minimum validator spec requirements. Move-based chains like Sui and Aptos leverage newer consensus algorithms like Hotstuff [2] and Mysticeti [3]. However, these approaches do not address the fundamental limitation of blockchains: they are subject to the trilemma where at a given security level, design approaches to achieve higher scalability necessarily reduce decentralization. Moreover, these chains operate on transparent global state where all data and computation are publicly visible, making privacy preservation difficult without additional cryptographic techniques.
The modular blockchain architecture, and layer-2 rollups in particular, offer a path forward by separating the system into specialized layers - consensus, data availability, execution, and settlement. This layered approach enables rollups to pursue performance optimization while relying on the base layer for security and censorship-resistance. At the time of writing, there are more than 50 production rollups on the market.
Layer-2s have made significant strides in scaling through optimized execution and specialized proof systems. However, they rely on centralized sequencers to order transactions, and RPC nodes to serve state to users. This architecture creates centralized bottlenecks during network congestion and points of failure during outages. They also require applications to constantly interact with their infrastructure even for operations that don't immediately need chain access or settlement.
Just as computer architecture relies on memory hierarchy that organizes complementary memory components based on their response times and capacities, we believe the World Computer vision requires complementary layers to achieve its full potential. Blockchains excel as decentralized hard drives, which from computer architecture’s point of view would sit at one end of the memory hierarchy. This observation compels us to ask:
What kind of systems would sit at the opposite end of the “memory hierarchy” for the World Computer?
To answer that question, we explore the properties a complementary layer would need to exhibit to complete the World Computing stack.
Picturing hard drive’s complement — RAM
RAM (random-access memory) complements hard drive in typical computer architecture: it is not suitable for long-term durable storage, but it offers far greater access speed. Let us imagine: if World Computer were to have a “RAM”, what properties would this new type of memory need to exhibit?
A world computer’s “RAM" should exhibit the following properties:
Ephemerality: Give up on global persistence.
Lock-free: Operate at much lower latency without resorting to centralized sequencers, by completely avoiding locks in distributed computation.
Many separate logs: Users interact only with their interested parts of this memory, unaffected by the activities in the rest of it.
Bypassing the blockchain trilemma: Computation occurs peer-to-peer between end user devices that are interested in it, relying on servers minimally. This allows computation to scale horizontally as interested users join and contribute their own computing resources, preserving maximal decentralization while scaling.
With these properties, a world computer’s “RAM” mirrors a RAM in common computer architecture: it is characterized by ephemerality (1), speed (2), “random access” (3), and closeness to end user compute (4).
This is a new type of memory that is fast and volatile, complementing blockchain's global persistent state.
The World Computer’s RAM
Topology is reimagining the world computer's architecture with RAM. To achieve this goal, we are developing two protocols that work in concert:
A new Internet protocol named Distributed Real-time Programs (DRP) [4]. DRP enables distributed applications to operate peer-to-peer between end users in real time, without the use of global consensus or centralized sequencers. State convergence is guaranteed by Byzantine Eventual Consistency [5], which contrasts and complements Byzantine consensus, the characteristic algorithm common to all public blockchains. Applications built on DRP benefit from real-time responsiveness, built-in privacy, and Byzantine fault tolerance.
An orchestration blockchain named Mandu. Mandu serves as an orchestrator between RAM's ephemeral, high-speed memory and blockchain's persistent global state. It provides efficient "gear-shifting" between these two domains, enabling applications to seamlessly transition between RAM and a wide variety of public blockchains. This orchestration function allows developers to leverage both domains to their maximum potential.
With these protocols, Topology establishes a memory hierarchy for the world computer, expanding its capabilities while preserving its decentralized nature. This architecture enables a new generation of applications that can dynamically balance performance, privacy, and persistence based on their specific needs.
How does RAM benefit base layers that are maximally decentralized?
RAM offers a scaling path complementary to rollups for decentralized base layers like Ethereum. Instead of relying on centralized servers for sequencing, RAM handles application state updates on end user devices, reducing load on the base layer and preserving censorship-resistant properties at the same time. This expands Ethereum's coverage of the scalability solution space by providing developers with options beyond the rollup-centric roadmap, particularly for applications requiring both real-time interaction and liveness guarantees.
How does RAM benefit a high-performance L2s?
High performance L2s still face inherent limitations from their reliance on servers for sequencing and indexing. RAM’s strength in client-side state management complements L2s at those limitations.
When applications handle state updates in RAM, they reduce dependency on L2 infrastructure and remain highly responsive during L2 network congestion or sequencer downtime. By managing interactions that don't require chain access, RAM can significantly reduce load on L2's RPC nodes. Combining RAM and L2s allows applications to benefit from both cheap settlement and uninterrupted speed.
How does RAM benefit high-performance L1s?
RAM preserves maximum decentralization by operating directly between end users without intermediaries, and when combined with these chains, applications can achieve both high performance and strong decentralization — using RAM for real-time interaction and L1 for fast settlement.
Furthermore, end-to-end encryption in RAM brings privacy preservation to these chains. By rendering states in cleartext form only in RAM — at end-user devices — applications can maintain self-custody of private activity logs while enjoying the high bandwidth and low latency of these chains for settlement in encrypted form. This hybrid approach allows developers to build for use cases that require both performance and confidentiality.
How does RAM benefit Web3 as a whole?
RAM serves as a technical and cultural bridge between Web3 and other decentralization communities, particularly the local-first community. Projects like Ink & Switch's local-first software initiative [6] are championing the shift of power from the cloud to end users by enabling real-time collaborative applications to operate smoothly even when users disconnect from the cloud. By sharing principles in real-time seamless collaboration and privacy preservation, RAM helps unite the decentralization communities toward a shared goal: to create an open and reliable multiplayer medium with built-in privacy and censorship-resistance.
RAM's ability to operate without requiring direct cryptocurrency interaction — when using DRP alone — also creates a more accessible entry point to Web3. Organizations and individuals who are interested in Web3 but hesitant about cryptocurrency can begin with RAM's no-middleman real-time computing capability, then gradually explore blockchain’s values as their needs evolve. We believe this opens up the ecosystem's reach beyond its current cryptocurrency-centric boundaries.
Reference
[4] paper.topology.gg
