Increase concurrency and DB pool sizes for scalability

Raised ThreadPoolExecutor worker limits from 50 to 200 in AI moderation and rule processing to improve OpenAI API concurrency. Increased SQLAlchemy pool_size to 50, pool_timeout to 60, max_overflow to 100, and DB_THREAD_POOL_WORKERS to 100 for better database performance under load.

Author: Bentlybro

app/services/ai/ai_moderator.py

@@ -325,9 +325,9 @@ def moderate_content(self, content, content_type='text', custom_prompt=None):
                         chunks.append(content[i:i + MAX_CHARS_PER_CHUNK])
 
                     # Process all chunks IN PARALLEL for maximum speed
-                    # Increased from 10 to 50 workers for better concurrency under load
+                    # Increased from 10 -> 50 -> 200 workers for OpenAI API concurrency
                     chunk_results = []
-                    with ThreadPoolExecutor(max_workers=min(len(chunks), 50)) as executor:
+                    with ThreadPoolExecutor(max_workers=min(len(chunks), 200)) as executor:
                         # Submit all chunks at once with context wrapper
                         future_to_chunk = {
                             executor.submit(self._context_wrapper, self._analyze_with_custom_prompt, chunk, custom_prompt): i
@@ -375,9 +375,9 @@ def moderate_content(self, content, content_type='text', custom_prompt=None):
                 chunks = self.split_text_into_chunks(content, max_content_tokens)
 
                 # Process all chunks IN PARALLEL for maximum speed
-                # Increased from 10 to 50 workers for better concurrency under load
+                # Increased from 10 -> 50 -> 200 workers for OpenAI API concurrency
                 chunk_results = []
-                with ThreadPoolExecutor(max_workers=min(len(chunks), 50)) as executor:
+                with ThreadPoolExecutor(max_workers=min(len(chunks), 200)) as executor:
                     # Submit all chunks at once with context wrapper
                     future_to_chunk = {
                         executor.submit(self._context_wrapper, self._run_enhanced_default_moderation, chunk): i

app/services/moderation/rule_processor.py

@@ -103,8 +103,8 @@ def process_single_ai_rule(rule):
                 return (rule.id, None)
 
         # Execute in parallel
-        # Increased from 10 to 50 workers for better concurrency under load
-        with ThreadPoolExecutor(max_workers=min(len(ai_rules), 50)) as executor:
+        # Increased from 10 -> 50 -> 200 workers for OpenAI API concurrency
+        with ThreadPoolExecutor(max_workers=min(len(ai_rules), 200)) as executor:
             futures = {executor.submit(
                 process_single_ai_rule, rule): rule for rule in ai_rules}
 

config/config.py

@@ -41,16 +41,16 @@ class Config:
 
     # SQLAlchemy connection pool configuration
     SQLALCHEMY_ENGINE_OPTIONS = {
-        'pool_size': 5,                    # Number of connections to maintain in pool
-        'pool_timeout': 30,                # Seconds to wait for connection from pool
+        'pool_size': 50,                   # Number of connections to maintain in pool (was 5)
+        'pool_timeout': 60,                # Seconds to wait for connection from pool (was 30)
         'pool_recycle': 1800,              # Seconds before recreating connection (30 min)
         'pool_pre_ping': True,             # Verify connections before use
-        'max_overflow': 10,                # Additional connections beyond pool_size
+        'max_overflow': 100,               # Additional connections beyond pool_size (was 10)
         'echo': bool(os.environ.get('SQL_DEBUG', False))  # SQL debugging via env var
     }
 
     # ThreadPoolExecutor configuration for async database operations
-    DB_THREAD_POOL_WORKERS = int(os.environ.get('DB_THREAD_POOL_WORKERS', '8'))
+    DB_THREAD_POOL_WORKERS = int(os.environ.get('DB_THREAD_POOL_WORKERS', '100'))
 
 
 class DevelopmentConfig(Config):