What methods have you used to evaluate and measure system efficiency during load testing?

To evaluate and measure the effectiveness of the system during load testing, I used the following methods:

• Performance metrics collection: response time, throughput (requests per second), CPU and memory usage.
• Server resource monitoring: tracking CPU, memory, disk, and network load using system tools (e.g., top, htop, sar) or specialized solutions (Prometheus, Grafana).
• Log analysis: identifying errors and bottlenecks through application logs.
• Using profilers: to identify "bottleneck" functions and operations.
• Creating graphs and reports: visualizing results for clear comparison of different scenarios.

For example, during load testing of a web service, I ran scenarios with varying numbers of concurrent users, measured average and maximum response times, error counts, and server load, then analyzed the data to optimize code and infrastructure.