Event-Driven KuCoin Triangular Arbitrage Framework in Async Rust

This is an async Rust project to implement zero-risk crypto trinagular arbitrage, explore technical feasiblity of generating passsive income (i.e. sleep to earn!).

How cyclic arbitrage works

Say we hold USDT, it checks all the listed crypto (e.g. ETH) that can trade against BTC and USDT, and compare the profit by either:

• Buy-Sell-Sell: buy ETH (sell USDT), sell ETH (buy BTC), sell BTC (buy USDT)
• Buy-Buy-Sell: buy BTC (sell USDT), buy ETH (sell BTC), sell ETH (buy USDT)

The above is triangular arbitrage, which is the simplest form of cyclic arbitrage. We can make more complex say finding the most profitable routes, or placing order at maker price for reduced fees.

Previous mistakes in Python

2 years ago I have made a python script that runs the triangular arbitrage in KuCoin, but it had several technical issues and ended up not following up.
https://github.com/kanekoshoyu/Kucoin-Triangular-Arbitrage

• Implementation with Python and REST polling was way too slow to obtain the valid arbitratge chances for execution.
• Generally the Python REST API caused quite high rates of communication error, which took extra time for resorting.
• It didn't count the actual size of the arbitrage order, which meant that the script kept buying some shitcoin and could not sell properly.
• It simply took the mean price instead of best bid/ask, which means it took maker positions for each of three actions in one arbitrage, and did not execute arbitrage promptly.

How to run example executables

1. Rename config_sample.toml as config.toml and set the API key with your own KuCoin API credentials. Configure the event monitor interval and the USD budget per each iteration of cyclic arbitrage.

[KuCoin Credentials]
api_key="YOUR_API_KEY"
secret_key="YOUR_SECRET_KEY"
passphrase="YOUR_PASSPHRASE"

[Behaviour]
# Performance monitor interval in seconds
monitor_interval_sec=120
# max amount of USD to use in a single cyclic arbitrage
usd_cyclic_arbitrage=100

2. At the root directory of the project(kucoin_arbiitrage), run the command below:

cargo run --bin event_triangular  

event_triangular is one of the example executables that implements [XXX-BTC, XXX-USDT, BTC-USDT] triangular arbitrage. There are other executables in the bin directory.

Overview

Code Structure

The project is split into these components:

Example Executables

• bin contains example executable codes. Some of them are for network testing purpose.

Internal Structure (Independent of Exchange APIs)

• model has internal generic data structures used for abstracted representations of markets. This should be independent of exchange APIs so that the the arbitrage strategy algorithm can be conducted across different exchanges.
• event has the events used to pass states and data passed across different components. It uses the internal model for the same reason.
• strategy has the implementations of arbitrage strategy algorithm. The algorithms are built upon internal model and event.
• monitor has the counter used to monitor MPS (message per seconds) for each broadcast channels, and a timers mapped globally by string for easy debug access.

Link to Exchange APIs (e.g. KuCoin)

• translator has the conversion of exchange API objects into internal models and vice versa. It uses traits and the traits are implemented per API models.
• broker has the tasks that runs API calls, and converts into internal data structure.

Event Pub/Sub with Tokio Broadcast

Event broadcasts empowers the modularity of tasks. Each async task communicates with eachother using events, pub/sub via tokio's broadcast. Here is the exmaple for event_triangular.rs

Task Pools with Tokio JoinSet

Tasks are grouped and spawned using JoinSets. We can either await for all the tasks to end with join! or await until a single task ends with select! or join_next. This provides full control over how we want to control these tasks. Here is the exmaple for task pools declared in core function of event_triangular.rs

When a task in taskpool returns, its result is received by join_next, which are received by core's select!. When an external signal is received, or core returns error, it gets detected by select! at the main and terminates the program.

Major Structural Improvements

• Use compiled Rust code for neat, efficient and proper code
• WebSocket subscription of real-time order books to get all the latest maker/taker ask/bid
• Only take a taker position based on the latest best-bid/ask price/size
• Implement both data bandwidth monitor and arbitrage performance monitor as tasks
• Abstraction of orderbook sync and arbitrage strategies using internal model and event
• Concurrency of syncing and strategy tasks

Feature Progress List

Deployment

Please refer to my another repo implementing service-level wrappers: chaiwala

Community

• Rust Questions: GitHub Repo Discussion
• Arbitrage Questions: Discord Server