Synchronous vs Asynchronous Methods - When to Use What (Python)¶
Understanding execution models in PyMT5: non-blocking streaming vs synchronous execution.
🎯 Quick Comparison¶
Python SDK uses asyncio for non-blocking network streams. Use async for on tick and trade generators. Synchronous wrappers are provided in Sugar for scripts and Jupyter.
| Aspect | Asynchronous Pattern | Synchronous Call |
|---|---|---|
| Thread Blocking | ❌ Non-blocking (high concurrency) | ✅ Blocks current thread |
| Throughput | ✅ Handles thousands of events/sec | ❌ Limited by thread pool |
| Real-Time Data | ✅ Perfect for tick & trade streams | ⚠️ Inefficient for streams |
| Simplicity | Requires async runtime awareness | Simple, linear execution |
| Recommended for | Automated bots, microservices, GUIs | CLI scripts, notebooks, one-offs |
🚀 When to Use Asynchronous Methods (Recommended)¶
1. Market Data Streaming¶
Market ticks arrive at microsecond intervals during peak sessions. Asynchronous handlers ensure zero tick drops without freezing your execution thread:
account = MT5Account(user=user, password=password, host=grpc_server)
await account.connect_by_server_name(server_name, "EURUSD", timeout=30)
summary = await account.account_summary()
print(f"Balance: {summary.account_balance}, Equity: {summary.account_equity}")
2. High-Frequency Order Execution¶
When operating across multiple currency pairs simultaneously, asynchronous dispatch allows your bot to send orders concurrently rather than sequentially.
💡 Summary¶
Always prefer asynchronous paradigms for production bots, multi-symbol trading, and background services. Use synchronous wrappers for quick setup scripts, testing, or exploratory analysis.