Скачать или смотреть BPA Platform best practices - optimize for performance, scalability, and maintainability

The Related Document can be found here: https://www.codelessplatforms.com/par...
You will need to be logged in with your Partner Details to access this file.

BPA Platform Best Practices

00:00 Introduction
00:26 Planning and Infrastructure
01:20 Performance and Optimisation
01:53 Task Design
02:35 Error Handling and Notifications
03:21 Maintenance and Analysis
03:41 Additional Recommendations
04:15 For More Information

Planning & Infrastructure
• Plan Ahead, Discuss performance expectations and infrastructure with your customer.

• Environment: BPA Platform runs as a Windows Service on Windows Server which can be cloud or on-premises. Ensure suitable hardware for scalability, particularly for cloud or VM environments.

• Requirements:
Align the hardware with the workload:
CPU: allocate Two dual-core CPUs for four concurrent tasks; and scale cores for higher concurrency.
Memory: Allocate 1GB RAM per concurrent task.
Disk: Use SSD for SQL Server databases; and avoid sharing servers with other systems unless the specifications are significantly upgraded.

Performance Optimization
• Batch Processing: Process records in groups instead of individual tasks to improve efficiency.
• Schedules & Triggers: Avoid frequent schedules (less than 5 minutes) unless essential. Schedule non-vital tasks during off-peak times.
• Concurrency: Configure threads to match task demand. Use “Run Sync” only when tasks must complete sequentially; otherwise, use “Queue.”

Task Design
• Modular Approach: Split tasks by function (e.g., customer and order integrations in separate tasks). Keep tasks concise and maintainable.
• Avoid Loops: Let BPA handle record processing loops. Use “Call Task (Queue)” for batch tasks instead of internal loops.
• Data Staging: Use a staging database to manage retries, minimize source system reprocessing, and isolate integration issues. This also promotes modular design and Task reuse in other solution combinations.

Error Handling & Notifications
• Error Resilience: Use flags (e.g., “Failed”/“Success”) in staging to identify and manage errors. Store failed transactions for reprocessing.
• Alerts: Configure email notifications for errors and task statuses. Use group email addresses for easy updates.
Web Services Integration
• API Efficiency: Use batch requests instead of multiple API calls. Consider API throttling limits and higher-capacity options if available.
• Data Validation: Check data types, schemas, and completeness when inferring from example data.

Maintenance & Analysis
• Event Log Analysis: Use BPA Platform’s 28-day event log to optimize task scheduling and load balancing.
• Housekeeping: Clear memory repositories regularly and validate task dependencies to prevent deadlocks.

Additional Recommendations
• Validate data at the source to reduce processing burdens.
• Avoid excessive task steps or combining multiple processes within a single task.
• Use BPA Platform connectors efficiently to minimize design complexity and enhance performance.

By following these practices, you can optimize BPA Platform solutions for performance, scalability, and maintainability.