Kubernetes Resource Requests and Limits are per-container settings that tell Kubernetes how much CPU and memory a workload needs and the maximum it can use. A request is the amount Kubernetes uses when scheduling a Pod onto a node, such as cpu: 500m or memory: 256Mi. A limit is the runtime cap, such as cpu: 1 or memory: 512Mi. If a container exceeds its memory limit, Kubernetes can terminate it with an out-of-memory error; if it exceeds its CPU limit, it is throttled. Teams use requests and limits to improve bin packing, reduce noisy-neighbor problems, control cost, and make Kubernetes infrastructure management more predictable, but values set too low can cause throttling or crashes, while values set too high can waste cluster capacity.
DevOps Glossary
Kubernetes Resource Requests and Limits
Kubernetes resource requests and limits set pod CPU and memory needs and caps, guiding scheduling and capping use.