Orderbook.trade uses a Hybrid Architecture designed to solve the "DEX Trilemma"
Providing CEX-level performance without sacrificing the non-custodial security of a DEX.
The Problem with Pure Chains
On-Chain AMMs: Too slow (block times) and expensive (gas on every action).
App Chains: Still limited by consensus overhead for every single order cancellation or modification.
Our Solution:
1. The Execution Layer (Off-Chain)
This is a high-performance Golang-Based Matching Engine hosted on our sequencer nodes.
Role: Maintains the Orderbook state, matches buy/sell orders, manages risk checks (margin), and sequences transactions.
Performance: Capable of processing 100,000+ orders per second with <5ms latency.
Safety: It cannot forge user signatures. It can only match valid orders signed by the user.
2. The Settlement Layer (On-Chain)
This is a set of Smart Contracts deployed on the target blockchain (Arbitrum, Solana, Monad, etc.).
Role: Holds user funds (Custody) and verifies the result of the off-chain matching.
Process: The engine submits a compressed "Batch" of trades. The contract verifies the math and updates user balances atomically.
3. The Security Model (Self-Custody)
Your funds are never held by the Orderbook.trade company server. They are strictly held in the Smart Contract.
Deposit: You call deposit() on the contract.
Trade: You sign an intent I want to swap A for B.
Withdraw: You can call withdraw() on the contract at any time.
What if the Orderbook goes down ?
We implement an "Escape Hatch" mechanism. If the sequencer goes offline or censors your withdrawal request, you can force a withdrawal directly from the Smart Contract after a timelock, guaranteeing you always have access to your funds.