Benchmark

Spegel is benchmarked using the benchmark. The tool creates pods in a cluster forcing image pulling, after all pods are running the tool with gather image pull metrics for each individual pod. The image is then updated causing a rolling update, which is configured to be 20% of the pods at a time. Once again the image pull time metric is gathered. These benchmarks are run as a suite for different image sizes and amount of layers. Refer to the benchmark results for details about how to reproduce the results.

Results

The results are compared to the baseline results which have been measured with the same setup but without Spegel running in the cluster.