Layer N's Nord Engine Achieves 100,000+ TPS on Closed Testnet
Layer N, today announced the results from its throughput testing, achieving a figure of 120,018 for transactions per second (TPS), 100x that of leading Ethereum scaling solutions. The testing was conducted on a closed testnet build that leverages Layer N’s new trading-optimized rollup engine - Nord Engine.
On Nord Engine, Layer N achieved 20,000 sustained TPS, 120,018 peak TPS, with EigenDA maintaining 3.0 megabytes per second (MBps). Nord Engine is the first app-specific rollup (ARN) introduced by Layer N within the StateNet, the term Layer N coined to represent a single shared state machine powered by a network of hundreds of custom and optimized rollups. Nord Engine was designed specifically to close the current gap between decentralized trading venues and centralized ones. Because of existing underlying compute and infrastructure constraints, decentralized trading venues have fewer features and worse latencies and throughput, typically by several orders of magnitude when compared to their centralized counterparts.
Nord Engine aims to not only close that gap, but also provide a product that can outperform the majority of centralized cryptocurrency exchanges in terms of performance. It is designed with a simple goal in mind: minimizing time-to-execution and maximizing overall throughput. Nord Engine can efficiently handle hundreds of thousands of transactions per second, high bursts of TPS without degraded performance, and sub-millisecond end-to-end latency. It also aims to enable something irreproducible in the centralized world: seamless composability with thousands of open and innovative on-chain financial applications through liquidity sharing, enabled by the StateNet architecture.
Dima Romanov, Layer N Co-Founder and CEO, said regarding today’s news: “After applications launch their Nord Engine instances, we plan on launching more optimized rollups including our own optimized version of the EVM, called N-EVM, that will be directly composable with Nord Engine. Exchanges deployed on Layer N will now be able to enable a whole ecosystem of developers to build hundreds of native integrations around them. Imagine if centralized exchanges we are all familiar with could allow developers to build applications directly on top of them. That’s what we are enabling.”
While the benchmarks were completed on a closed testnet instance, Layer N plans to open the testnet build to the public in the coming weeks. Follow Layer N on Twitter at @LayerN_Official to stay updated on the details of their public testnet launch.
More articles
Introducing Atomics on N1
Atomics lets traders bundle multiple actions into a single transaction. On 01, one atomic call can include up to 10 coordinated place and cancel actions. The system executes them in order and checks validity and risk after each step. If any step fails, the whole transaction is rejected and nothing lands in a partial state. On a live perp orderbook, that means a basket entry, hedge adjustment, or quote refresh can be sent as one instruction set instead of being broken into separate transactions while the book keeps moving.
L1's are not infra (or they shouldn't be)
L1s keep getting the story wrong. They operate like infrastructure companies, charging per transaction like AWS charges per compute cycle, when they're actually financial networks.
Multiple Concurrent Proposers is not a scaling solution
Writing this to address a common misunderstanding. Solana's MCP proposal does not solve scaling. It doesn't even claim to! Yet this misunderstanding persists. MCP solves a fundamentally different problem; crucial, but distinct from horizontal scaling, which is what centralized ledgers with bridging attempt to solve.