Skip to content

Polkadex Design Principles

Gautham edited this page Oct 7, 2020 · 2 revisions

Polkadex Design Principles

The Polkadex Trading engine is designed to have only essential logic on-chain, namely.

  • Orderbook
  • User Assets management
  • Market Data Collection

Everything else, like the aggregation of Market Data, Technical Analysis of market data, Order History management, are made off-chain. It can be achieved on the client-side or cloud. Polkadex assumes that the majority of trades are market making, so also tries to optimize market-making trades. The trades are internally stored in a First-In-First-Out queue grouped together by Price and Market. The different Price levels of the market are stored in a LinkedList, which allows Polkadex to have minimum reads and writes to blockchain state. Blockchain storage is read-only if it is sure that it is able to consume orders or make a new order. It enables us to minimize the time needed for execution.

More technical information about the Polkadex Trading engine will be added here.

Clone this wiki locally