Plan and optimize Kubernetes resource requests and limits. Calculate per-container, per-pod, and per-deployment totals with QoS and YAML output.
Container Configuration
CPU Resources
Memory Resources
You might also find these calculators useful
Calculate optimal K8s node sizes and cluster configuration
Calculate optimal Kubernetes pod replicas for your workload
Calculate HPA scaling triggers and threshold zones for Kubernetes
How many users one server holds, and what limits it
Accurately calculate CPU and memory resources for your Kubernetes deployments. This calculator helps you determine appropriate request and limit values for containers, pods, and entire deployments while ensuring optimal Quality of Service (QoS) class assignment.
In Kubernetes, CPU and memory are allocated to containers through resource requests and limits. A request is the amount the scheduler reserves when placing a Pod on a node, so total requests across Pods cannot exceed a node's allocatable capacity; a limit is the ceiling a container may consume at runtime. CPU is measured in cores or millicores (1000m = 1 core) and is compressible (throttled when exceeded), while memory is measured in bytes (Mi/Gi) and is incompressible, so exceeding a memory limit triggers an OOMKill. Right-sizing means setting requests close to real usage so nodes are packed efficiently without starving or overcommitting workloads.
A DevOps engineer sets CPU and memory requests for a pod based on the workload it needs to handle
A platform team calculates how many nodes are needed to run all scheduled workloads without over-committing resources
An engineer estimates resource limits to configure horizontal pod autoscaling thresholds correctly
An SRE checks whether reducing over-allocated requests would free node capacity and lower cluster cost
Set appropriate requests to ensure your pods get scheduled and receive guaranteed resources.
Configure proper limits to prevent runaway containers from impacting other workloads.
Right-size resources to maximize cluster efficiency without over-provisioning.
Understand how your resource configuration affects pod priority during node pressure.
Requests are guaranteed resources used for scheduling decisions. Limits are the maximum resources a container can use. A container can exceed its request (if available) but never its limit.
Guaranteed: requests equal limits for all resources. Burstable: at least one request or limit set, but not equal. BestEffort: no requests or limits set. During node pressure, BestEffort pods are evicted first.
Not necessarily. CPU is compressible, so setting higher limits allows bursting. However, for predictable performance, some prefer matching request and limit for Guaranteed QoS.
Unlike CPU which gets throttled, exceeding memory limit triggers OOMKill (Out of Memory Kill). The container is terminated and potentially restarted based on restart policy.
Start by profiling your application under load. Set requests to the typical usage and limits to handle peak load. Use tools like kubectl top or Prometheus to monitor actual consumption.