multithreading#

Python thread safe operations

Quick example from the official Python documentation about thread safety in Python:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

L.append(x)
L1.extend(L2)
x = L[i]
x = L.pop()
L1[i:j] = L2
L.sort()
x = y
x.field = y
D[x] = y
D1.update(D2)
D.keys()

1
2
3
4

i = i+1
L.append(L[-1])
L[i] = L[j]
D[x] = D[x] + 1

It's important to understand that Python, due to its Global Interpreter Lock (GIL), can only switch between threads between bytecode instructions. The frequency of these switches can be adjusted using sys.setswitchinterval(). This ensures that within a single bytecode instruction, Python will not switch threads, making the operation atomic (thread-safe). For a deeper dive into this topic, you can read this discussion on atomic and thread-safe operations in Python.

Using Python SQLAlchemy session in concurrent threads or tasks

SQLAlchemy DB session is not thread safe for both sync and async session. AsyncSession is only a thin proxy on top of a Session

The concurrency model for SQLAlchemy's Session and AsyncSession is therefore Session per thread, AsyncSession per task.

The best way to ensure this use is by using the standard context manager pattern locally within the top level Python function that is inside the thread or task, which will ensure the lifespan of the Session or AsyncSession is maintained within a local scope.

For applications that benefit from having a "global" Session where it's not an option to pass the Session object to specific functions and methods which require it, the scoped_session approach can provide for a "thread local" Session object; see the section Contextual/Thread-local Sessions for background. Within the asyncio context, the async_scoped_session object is the asyncio analogue for scoped_session, however is more challenging to configure as it requires a custom "context" function.