A harmful workload generator for PostgreSQL that simulates various database stress scenarios for testing purposes.
Noisia is a harmful workload generator for PostgreSQL that simulates various database stress scenarios such as deadlocks, connection exhaustion, and transaction blocking. It helps developers and database administrators test system resilience, monitor performance under duress, and validate failover mechanisms in controlled environments.
Database administrators, DevOps engineers, and backend developers who need to test PostgreSQL database resilience, performance under stress, and failover mechanisms.
Noisia provides a specialized, open-source tool for simulating real-world database stressors, allowing teams to proactively identify weaknesses and improve system robustness without risking production data.
Harmful workload generator for PostgreSQL
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Targets unique failure modes like deadlocks and connection exhaustion, as listed in the README's workload types, enabling precise resilience testing for PostgreSQL databases.
Provides pre-built Docker images with example commands in the README, allowing quick setup and execution without complex installation steps.
Can be imported as a Go package for custom workflows, demonstrated in the code example, facilitating automated testing and integration into existing systems.
Includes a detailed table in the README specifying which workloads cause performance degradation, helping users anticipate and monitor effects.
The strong disclaimer warns that reckless usage can cause database problems like connection exhaustion or data bloat, requiring expert knowledge for safe operation.
Exclusively designed for PostgreSQL, making it ineffective for teams managing multi-database environments or other SQL databases like MySQL or SQLite.
Offers only predefined stress scenarios with minimal options for tailoring to specific application logic, as indicated by the fixed workload list and reliance on built-in help.