Kubernetes CPU and Memory Calculator
Convert Kubernetes CPU and memory quantities (500m, 1.5, 512Mi, 2G), add up requests across replicas, see each pod's QoS class and estimate the nodes needed.
Free · No sign-up · Your inputs stay in your browser
Input
Convert and add up Kubernetes CPU and memory
Output
1.75 cores, 1.70 GiB requested
| Workload | CPU req / lim | Memory req / lim | QoS class |
|---|---|---|---|
| web | 250m / 1000m | 256 MiB / 512 MiB | Burstable |
| worker | 500m | 488 MiB / 1024 MiB | Burstable |
Nodes needed (requests only)
1
at 4 vCPU / 16 GiB each
worker: 512M is decimal (488.3 MiB). Mi and Gi are the powers of 1024 most people mean.
The scheduler places pods by their requests, never their limits. A container over its CPU limit is throttled; one over its memory limit is OOM-killed. Node count assumes about 10% of CPU and 15% of memory are kept by the kubelet and system pods; check kubectl describe node for the real allocatable.
Runs in this browser. Nothing you paste is sent to a server.
About this tool
Method, assumptions, practical next steps, and limits.
Requests schedule, limits enforce
The scheduler places a pod on a node with enough unrequested capacity; limits are not considered. At runtime a container over its CPU limit is throttled, and one over its memory limit is killed with OOMKilled. A pod whose requests equal its limits for every container gets the Guaranteed QoS class and is the last evicted under pressure. See Resource Management for Pods and Containers.
M is not Mi
512M is 512,000,000 bytes and 512Mi is 536,870,912. Mixing them is harmless for requests and confusing for limits, where a container sized for 512Mi can be killed at 488Mi.