Current threading model is too slow and not concurrent enough, in part due to Python's global interpreter lock - migrate to full multiprocessing solution: hold Folia document in one thread, collect FQL queries, execute them sequentially in that thread. Have other threads communicate concurrently with the various servers.
proycon
commented
9 years ago
Current threading model is too slow and not concurrent enough, in part due to Python's global interpreter lock - migrate to full multiprocessing solution: hold Folia document in one thread, collect FQL queries, execute them sequentially in that thread. Have other threads communicate concurrently with the various servers.