remsky
Excuse me, how does this utilize FastAPI and other technologies to solve the Python Global Interpreter Lock (GIL) issue, and how can it achieve concurrency?
Excuse me, how does this utilize FastAPI and other technologies to solve the Python Global Interpreter Lock (GIL) issue, and how can it achieve concurrency?
How is this an issue and it achieves concurrency because every request is run in a different thread ( this is default fast-api behaviour). As far as I know this does not solve GIL
How is this an issue and it achieves concurrency because every request is run in a different thread ( this is default fast-api behaviour). As far as I know this does not solve GIL
When the model runs on a GPU, it is a computational task. Since Python will not release the Global Interpreter Lock (GIL), it is not truly multithreaded. So, how can concurrency be achieved? Only in I/O-bound tasks will the GIL be released briefly, enabling a short-lived multithreading state.
So that means the clients calling the fastapi are queued and processes one by one due to the GIL? If want to improve the performance for support massive clients call, what is the suggestion? Does this docker image support Python multiple process tech(i assume it's a solution)?
