toolkit

K8s Resource Calculator

Runs in browser

Calculate CPU & memory requests/limits

Quick Presets

CPU Resources

0.1 cores

0.5 cores

0 Request: 100m Limit: 500m 2000m

Memory Resources

0.13 Gi

0.50 Gi

0 Request: 128Mi Limit: 512Mi 2Gi

Node Capacity (for density calc)

How to Use

Set CPU and Memory requests/limits to see analysis.

You will see:

  • QoS Class (Guaranteed, etc.)
  • OOM & Throttling Risk
  • Pod Density Analysis
Analysis

QoS Class

Burstable

Medium priority. Has some requests/limits set but not equal.

CPU Throttle Risk

low

Burst: 5.0x request

Memory OOM Risk

low

Burst: 4.0x request

Pods per Node

40

Limited by CPU

Overcommit

Yes

CPU, Memory

Generated YAML

resources:
  requests:
    cpu: "100m"
    memory: "128Mi"
  limits:
    cpu: "500m"
    memory: "512Mi"

Understanding Resources

Request: Guaranteed resources. Used for scheduling.

Limit: Maximum allowed. Exceeding CPU causes throttling, exceeding memory causes OOM kill.

Throttling: When CPU usage exceeds limit, container is throttled (slowed down).

OOM Kill: When memory exceeds limit, container is killed and restarted.

Related Tools

Kubernetes

K8s Probe Generator

Simulators

Pod Eviction

Simulators

K8s Rollout

Guides

How Autoscaling Works

About Kubernetes Resource Management

Kubernetes resource requests and limits define how much CPU and memory containers can use. Proper configuration ensures efficient cluster utilization, prevents resource starvation, and enables effective autoscaling.

Requests vs Limits

Requests

  • Guaranteed resources for scheduling
  • Pod won't schedule without available resources
  • Used for scheduling decisions

Limits

  • Maximum resources allowed
  • CPU throttled if exceeded
  • OOMKilled if memory exceeded

Resource Units

CPU

1
= 1 vCPU/core
500m
= 0.5 CPU (millicores)
100m
= 0.1 CPU = 10%

Memory

128Mi
= 128 Mebibytes
1Gi
= 1 Gibibyte
512M
= 512 Megabytes (decimal)

QoS Classes

Guaranteed

Requests = Limits for all containers. Highest priority, last to be evicted.

Burstable

Requests < Limits. Can burst above requests if resources available.

BestEffort

No requests/limits set. Lowest priority, first to be evicted under pressure.

💡 Pro Tips

  • Set requests based on actual usage from metrics
  • Set limits 20-50% above requests for burst capacity
  • Use Vertical Pod Autoscaler (VPA) to right-size resources
  • Monitor OOMKilled events and CPU throttling
  • Consider using LimitRanges and ResourceQuotas for governance

References & Further Reading