Seeing is knowing, measuring CPU throttling in containerized environments
Containers and orchestration overcame a lot of challenges
Among the main benefits they allowed for increased density of applications deployed on each host. In the current scenario careful resource management becomes critical, and brings back some old problems, such as noisy neighbors, on the table.
One of the key phenomena that affects application performance in this context is throttling. We will dive into it, understanding how we can use it as a perfect indicator for saturation. Measuring throttling and its effect both to consumption metrics and end user impact is critical but still hard.
We will introduce what is throttling and show how popular open source and enterprise monitoring solutions address this problem. We will also try to identify new metrics to better understand when throttling is happening. Finally, by the use of benchmark Java applications, we will provide guidelines to effectively understand this important phenomenon and its impact on end users.
Edoardo is a performance testing consultant at Moviri, helping enterprise customers to achieve better performance, scalability, and resiliency.Despite being only 26 years old, he already has 3 years of experience in the Performance Engineering field with a focus on java applications in containerized environments. He actively designs and conducts performance testing activities in complex enterprise environments taking care of all aspects from workload generation to monitoring and inspection of performance metrics. Passionate about the IT industry and the Performance Engineering field, he loves to discover and try different tools and methodologies to discuss with the testing teams.