I am trying to simulate a server accepting requests (node IDs for a graph) for which it performs inference using a trained GNN (Graph Neural Network) model. Every time a request arrives, the server should launch an asynchronous thread to run inferencing for that node. To do so, I have used ThreadPoolExecutor from concurrent.futures. As soon as I run the script for more than 1 worker, I get an Aborted (core dumped) error and the process is killed. I have attached the code for reference. Can anyone point out how to get around this issue and what exactly am I doing wrong?

I am trying to simulate a server accepting requests (node IDs for a graph) for which it performs inference using a trained GNN (Graph Neural Network) model. Every time a request arrives, the server should launch an asynchronous thread to run inferencing for that node. To do so, I have used ThreadPoolExecutor from concurrent.futures. As soon as I run the script for more than 1 worker, I get an Aborted (core dumped) error and the process is killed. I have attached the code for reference. Can anyone point out how to get around this issue and what exactly am I doing wrong?

I am trying to simulate a server accepting requests (node IDs for a graph) for which it performs inference using a trained GNN (Graph Neural Network) model. Every time a request arrives, the server should launch an asynchronous thread to run inferencing for that node. To do so, I have used ThreadPoolExecutor from concurrent.futures. As soon as I run the script for more than 1 worker, I get an Aborted (core dumped) error and the process is killed. I have attached the code for reference. Can anyone point out how to get around this issue and what exactly am I doing wrong?