2025-02-13 –, RoboCon
Upgrade your PostgreSQL database with confidence! Learn how we used RF Swarm to simulate peak traffic and run performance tests before and after upgrading to PostgreSQL 15.4 on AWS. Discover how we ensured a smooth transition, improved query efficiency, and minimized risks—all with seamless integration of existing tools.
Before upgrading our production database to PostgreSQL 15.4 on AWS, we conducted performance testing to ensure the system could handle production-like conditions. We began by running baseline tests on the current setup to understand its performance under load, measuring key metrics such as CPU utilization, read latency, and freeable memory. This provided a reference point for comparison after the upgrade.
To simulate real-world usage, we used RF Swarm, a distributed load-testing tool integrated with the Robot Framework (RF). RF Swarm allowed us to simulate peak traffic, replicating the most demanding times of our application usage. Because RF Swarm works seamlessly with existing Robot Framework tests, we were able to extend our automated tests for performance testing with minimal effort. This saved time and allowed for efficient test execution.
Once the baseline was established, we upgraded to PostgreSQL 15.4, which introduced several performance optimizations, including improved indexing and query planning. After the upgrade, we reran the same RF Swarm tests to compare results with the baseline. The upgraded database performed well, with improved query efficiency, particularly for complex queries involving large datasets. CPU utilization decreased slightly during peak load, indicating that PostgreSQL 15.4 handled requests more efficiently. Read latency remained stable, and no significant memory management issues were detected.
The results validated that the upgraded PostgreSQL version could maintain the performance levels required for our production environment, while also introducing improvements in efficiency. The seamless integration of RF Swarm and Robot Framework allowed us to conduct comprehensive tests without needing to rework our existing setup.
In conclusion, by conducting thorough pre- and post-upgrade performance tests using RF Swarm and Robot Framework, we ensured a smooth and successful transition to PostgreSQL 15.4. The new version not only maintained but enhanced the performance of our application, minimizing the risks associated with the upgrade and giving us confidence that the system would perform reliably under peak load conditions in the AWS environment.
In-Person
Categorize / Tags:Performance Testing
Lessons Learned:In this talk, participants will gain practical insights into the process of upgrading a PostgreSQL database while ensuring optimal performance. Here’s what attendees can expect to learn:
Key Takeaways
Performance Testing Methodologies: Participants will understand the importance of establishing performance baselines before any database upgrade and learn how to effectively conduct these tests.
Hands-On with RF Swarm: Attendees will gain practical experience with RF Swarm for load testing, including how to set up simulations of peak traffic to evaluate database performance accurately.
Integration with Robot Framework: Learn how to leverage existing Robot Framework tests for performance testing, streamlining the process of creating a comprehensive testing strategy.
Understanding Key Performance Metrics: Participants will become familiar with critical metrics like CPU utilization, read latency, and freeable memory, and learn how to interpret these metrics to assess database health.
Best Practices for Database Upgrades: We will share lessons learned and best practices for minimizing risks during the upgrade process, including tips on query optimization and efficient resource management.
Real-World Case Studies: Participants will hear real-world examples and “war stories” from our upgrade experience, providing insights into challenges faced and solutions implemented.
Detailed Schedule (Approx. 45 minutes)
Introduction (5 minutes): Overview of the talk and importance of performance testing before database upgrades.
Establishing Baselines (10 minutes): Techniques for measuring current performance and setting up tests.
Using RF Swarm for Load Testing (10 minutes): Live demonstration on how to set up RF Swarm, including simulating peak usage scenarios.
Interpreting Performance Metrics (5 minutes): Understanding and analyzing key metrics to assess performance.
Best Practices and Lessons Learned (10 minutes): Sharing insights from our upgrade process, including what worked well and what to avoid.
Q&A Session (5 minutes): Open floor for questions, allowing participants to seek clarification or deeper insights.
Value to Participants
By the end of the session, attendees will walk away with actionable knowledge and techniques to confidently approach their own PostgreSQL upgrades, armed with the skills to conduct thorough performance testing and make data-driven decisions to optimize their database environments.
This talk is intended for a targeted audience comprising database administrators, Test engineers, DevOps engineers, and IT professionals involved in database management and application development. Specifically, it will benefit:
Database Administrators: Those responsible for maintaining and optimizing database performance who seek to improve their understanding of PostgreSQL upgrades and performance testing methodologies.
Test Engineers: Test engineers looking to gain insights into how performance testing can be done using Robot Framework.
DevOps Engineers: Professionals focused on automation and continuous integration/deployment who want to learn how to integrate performance testing into their workflows using tools like RF Swarm and Robot Framework.
The content will be technical, making it most suitable for participants with a foundational knowledge of databases and performance testing.
Is this suitable for ..?:Beginner RF User, Intermediate RF User, Advanced RF User
Test Engineer | DevSecOps at Midaxo