19TUNING QUALITY OF SERVICE
Ideally, our applications would use minimal or highly predictable processing, memory, storage, and network resources. In the real world, though, applications are “bursty,” with changes in load driven by user demand, large amounts of data, or complex processing. In a Kubernetes cluster, where application components are deployed dynamically to various nodes in the cluster, uneven distribution of load across those nodes can cause performance bottlenecks.
From an application architecture standpoint, the more we can make the application components small and scalable, the more we can evenly distribute load across the cluster. ...
Get The Book of Kubernetes now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
