Ep. 14 Google Kubernetes Engine(GKE) Explained in 15 minutes

Hello Everyone,

In this video, You will learn all about GKE (Google Kubernetes Engine). I have tried to cover all the important topics from the associate cloud engineer exam perspective along with a detailed hands-on demo.

Below topics covered in this video in this 15 minutes video:
- Introduction
- Why do we need Kubernetes?
- What is GKE?
- How to create a GKE cluster
- Standard vs auto-pilot GKE cluster
- What is a node pool in GKE?
- Cluster autoscaling in GKE
- Vertical and horizontal pod autoscaling in GKE
- Node auto-repair vs node auto-upgrade
- How to interact with the GKE cluster
- How to deploy a simple application to GKE
- How to set default cluster in GKE
- gkectl vs kubectl commands
- knowledge checks for the associate cloud engineer certification exam.

𝗖𝗢𝗠𝗣𝗟𝗘𝗧𝗘 𝗧𝗨𝗧𝗢𝗥𝗜𝗔𝗟 𝗦𝗘𝗥𝗜𝗘𝗦 👌 It’s 🅵🆁🅴🅴 ⛳
Namaste Google Cloud | GCP Tutorial for beginners, a complete playlist for Associate Cloud Engineer Certification Exam:

DevOps Tutorial for beginners

AWS tutorial for beginners

Azure Data fundamentals DP 900 full course

Jenkins Tutorial for beginners

DevOps and Cloud Podcasts:

Subscribe to our channel to get notified about the latest videos.

Connect with me 👋
🌎LINKEDIN: ►
🐦TWITTER: ►

Join our learning communities 👋

References 📚:

References 📚:

#gke #techtutorialswithpiyush #googlecloud

[TechTutorialswithPiyush]

🔥 Check out our GCP playlist Namaste Google Cloud - Google Cloud Tutorial for beginners

[kiransomwanshi]

Hi Piyush, Thanks for awesome video, very well explained with actual demo. Please refer below answers to knowledge check questions.

1) gcloud gives you the ability to provision and deprovision resources, whereas kubectl gives you the ability to use the clusters once provisioned.
2) False. Because GKE manages your node with minimal configurations required.
3) When we need to create a similar infrastructure(machine type, machine family, boot disk type) to multiple compute engine instances.
4) False. When demand is high, the cluster autoscaler adds nodes to the node pool not to cluster.
5) Horizontal scaling means that the response to increased load is to deploy more Pods. This is different from vertical scaling, which for Kubernetes would mean assigning more resources (for example: memory or CPU) to the Pods that are already running for the workload.
6) gcloud config set container/cluster <your cluster name>
7) GCP recommend that you opt into a release channel and enable node auto upgrades to help reduce the operational burden involved with upgrading GKE versions. However, when manually upgrading, GCP recommend planning to upgrade no later than every six months to gain access to new features and remain on a supported version.
8) maxSurge: The number of pods that can be created above the desired number of pods during an update. maxUnavailable: The number of pods that can be unavailable during the update process.
9) Public cluster – to configure access from public network to cluster’s workload. Private cluster – to assign internal IP addresses to pod and nodes. This isolate cluster workloads from public networks.

[naskar]

First of all thanks for the video piyush. Like other videos in the playlist I expected it be beginner friendly but the initial couple of minutes and the last few minutes were rushed I felt. but that is me... not sure how others took it. Like for some one who doesn't know basics of kubernetes some key words felt like a foreign land.

Knowledge Check:
1. gcloud when communicating with google resources and kubectl when using cluster plane or cluster related components
2. false since google manages the cluster and does all the work for you
3. When similar resources are expected?
4. false
5. Horizontal Pod scaling -> increase the number of pods, Vertical scaling -> increase the definition (like memory, CPU etc) of pod.
6. gcloud config set container/cluster cluster_name
7. select release channel for control plane version
8. max surcharge - number of pods to deploy when there is a max load, max unavailable - max number that can be free or unavailable
9. Public - cluster accessible to public, private - only with in GCP (accessible via internal IP addresses)

[KARADVAIBHAVPRAKASH]

I appreciate you sharing this video and I know it will make your future ones more interesting also your expression shows that you are becoming more engaging.

Q.1 gcloud cmd when we use for interaction with k8s like delete update & configure other hand kubectl cmd is only interacts inside the cluster provision
Q.2 False because of autopilot all GKE manages your pods and there is minimal configuration
Q.3 When suddenly traffic increase
Q.4 True
Q.5 Horizontal pod are basically increase or decrease the pods based on CPU utilization, on the other hand, Vertical pod autoscaling scale the resource to allocate to each pods and adjust the CPU and memory request.
Q.6 gcloud config set clusere/container name
Q.7 GCP have to release the channel and also they recommended opting the for node auto upgrade. For manualits give you broad access so depends on you when you want to upgrade.
Q.8 Max-Surge The maximum number of additional pods that can be created during an update, above the target number of pods. Max-Unavailable The maximum number of pods that may be unavailable while an update is being applied.
Q.9 Public GKE cluster are accessible from anywhere on the internet. Private GKE cluster node use internal IP only this provide additional layer of security.

[DharmendraYadav-jnyp]

Great video Piyush ❤️Loved it keep it up