Optimize deployment with shared memory between threads instead of new workers

Dockerfile

@@ -12,7 +12,11 @@ COPY --chown=user . .
 
 EXPOSE 7860
 
-CMD ["gunicorn", "app:server", "--bind", "0.0.0.0:7860", "--workers", "4"]
+# Optimization: Baseline
+# CMD ["gunicorn", "app:server", "--bind", "0.0.0.0:7860", "--workers", "4"]
+
+# Optimization: Shared memory with many threads
+CMD ["gunicorn", "app:server", "--bind", "0.0.0.0:7860", "--workers", "1", "--threads", "4", "--timeout", "120"]
 
 # Option 1: Point directly to the Dash app's server. -- does not work for hf spaces
 # CMD ["gunicorn", "app:app.server", "--bind", "0.0.0.0:7860", "--workers", "4"]