32. Kubernetes Resource Requests & Limits Explained + Demo

Kubernetes Resource Requests & Limits are key for stable workloads, preventing CPU throttling, OOM kills, and performance issues—so in this video, I break down how Kubernetes CPU/Memory Requests, Limits, QoS classes, and Scheduling actually work, followed by a simple live demo to help you understand Pod behavior in real clusters. You’ll learn how Kubernetes decides where to schedule Pods, why resource throttling happens, how to avoid crashes, and how to apply best practices for production environments using kubectl. ✔️ What are Requests & Limits? ✔️ How Kubernetes Scheduling works ✔️ QoS Classes (Guaranteed, Burstable, BestEffort) ✔️ Real Pod behavior demo ✔️ Tips to avoid throttling & OOM 🔥 Want more DevOps & Kubernetes tutorials? Subscribe for more!

👉 If this helps, please like 👍, subscribe, and share to support the channel!

🎓 Continue Learning with These Playlists :

🐧 Linux for Beginners (Start from scratch):
👉 [Watch the Playlist](   • Linux for Beginners 🐧  )

💻 Linux Command Line Tutorials (Terminal-focused skills):
👉 [Explore Commands](   • Linux Command  )

🚀 Kubernetes Series for Admins and Devs :
👉 [Start Learning Kubernetes](   • Kubernetes  )

📌 Commands Used in This Video
kubectl top pod
kubectl apply -f resource-demo.yaml
kubectl describe pod
kubectl delete pod resource-demo
kubectl describe node minikube | grep -A5 Capacity
minikube start --cpus=4 --memory=4096 --driver=docker
kubectl get pods -n kube-system | grep metrics
minikube addons enable metrics-server
Check QoS
kubectl get pod qos-guaranteed qos-burstable qos-besteffort -o=custom-columns=NAME:.metadata.name,QOS:.status.qosClass

📚 Chapters
0:00 🎬 Intro to Requests and Limits
0:53 ❓ Why Resource Management Matters
2:01 📝 Resource Requests (The Minimum)
3:09 🚫 Resource Limits (The Maximum)
5:13 🆚 Request vs. Limit (Scheduling vs. Runtime)
6:33 📐 CPU and Memory Units Explained
7:58 💣 Cluster Chaos: The Danger of No Limits
9:29 🗺️ How Kubernetes Schedules Pods
10:51 🎖️ Quality of Service (QoS) Classes
12:14 🍕 Real-World Analogy (The Restaurant)
13:43 💻 Demo: Scheduling Plan
14:24 ⚙️ Demo: Setup and Normal Scheduling
15:38 📈 Demo: Check Node Capacity
16:35 📝 Info: Commands to Reduce CPU
18:57 💥 Demo: Force Pod to Pending State
20:25 🔎 Demo: Describe Pending Pod (Insufficient CPU)
20:55 ✅ Demo: Revert Request and Schedule Pod
22:15 📊 Demo: Check Pod Usage (kubectl top)
23:51 💡 Info: Metrics Server Requirement
25:15 🧹 Demo: Cleanup 1

💻 Demo: CPU Throttling
25:32 ⚡ Demo: CPU Throttling Setup
26:08 🚀 Demo: Apply Pod with Limit
26:26 ⏳ Demo: Generate Max CPU Load
27:07 📈 Demo: Observe Throttling with kubectl top
28:48 🛑 Demo: Stop Load and Cleanup

💻 Demo: OOM (Out-of-Memory) Kill
29:48 💀 Demo: OOM Kill Setup
30:05 🚀 Demo: Apply Pod with Low Memory Limit
31:02 🗑️ Demo: Observe OOM Kill and Crash Loop
32:19 📝 Demo: Describe Pod to Confirm OOM Reason
33:26 🧹 Demo: Cleanup 2

💻 Demo: QoS Classes
33:48 🎖️ Demo: Guaranteed QoS Pod
34:33 🥈 Demo: Burstable QoS Pod
35:09 🥉 Demo: Best Effort QoS Pod
35:36 🚀 Demo: Deploy All QoS Pods
36:38 🔎 Demo: Verify QoS Classes via Command
37:54 🧹 Demo: Cleanup 3
38:44 🏁 Final Recap and Best Practices
39:41 👋 Outro

#Kubernetes #KubernetesTutorial #KubernetesScheduling