What changes occur with threads during significant server system overload?

Under significant server system overload with threads in Python, the following changes may occur:

• Increased latency: threads often block, waiting for resources to be released, leading to delays.
• Increased context switches: with a large number of threads, the OS spends more time switching between them, reducing overall performance.
• Race conditions and deadlocks: under high load, synchronization errors become more likely.
• Performance loss due to GIL (in CPython): in multi-threaded Python applications with CPU load, threads do not execute in parallel because of the global interpreter lock, which worsens under load.

As a result, the system may experience performance degradation, delays, and instability in thread operation.