fury licence pyversions Publish Build Docs

A very simple asynchronous wrapper that allows you to get access to the Oracle database in asyncio programs.

Easy to use , buy may not the best practice for efficiency concern.


  • cx_Oracle >= 8.1.0 (Take into consideration that author of cx_Oracle said he’s trying to implement asyncio support , APIs maybe change in future version. Switch to 8.1.0 if there’s something wrong makes it not gonna work.)
  • ThreadPoolExecutorPlus >= 0.2.0


pip install cx_Oracle_async


  • Nearly all the same as aiomysql in asynchronous operational approach , with limited cx_Oracle feature support.
  • No automaticly date format transition built-in.
  • AQ feature added , check docs here for further information.
  • You can modify some of the connection properties simply like you’re using cx_Oracle.
  • You can do basic insert / select / delete etc.
  • If you’re connecting to database which is on a different machine from python process , you need to install oracle client module in order to use this library. Check cx-Oracle’s installation guide for further information.



query type asynchronous multithreading synchronous multithreading synchronous single thread
fast single line query 6259.80 q/s 28906.93 q/s 14805.61 q/s
single line insertion 1341.88 q/s 1898 q/s 1685.17 q/s

/* Test platform: */
AMD Ryzen 3700x
Windows 10 LTSC
Oracle 19c
You can find performance test codes here.


Before running examples , make sure you’ve already installed a Oracle Client on your machine.

import asyncio
import cx_Oracle_async

async def main():
    oracle_pool = await cx_Oracle_async.create_pool(
        min = 2,
        max = 4,

    async with oracle_pool.acquire() as connection:
        async with connection.cursor() as cursor:
            await cursor.execute("SELECT * FROM V$SESSION")
            print(await cursor.fetchall())

    await oracle_pool.close()

if __name__ == '__main__':