Basics - k8s
Kubernetes is a platform
It can work with any type of container, not only docker
At its core, its gives you a platform that can
1) Scale easily
2) Easy to do monitoring
3) Means to do deployments
Master node is the kubernetes Master
You decide how to do stuff
You talk to Kubernetes and tell it what kind of image you want to use to create a container from and give it other critereon
It creates a deployment
Deployment can say a bunch of different things
i need a certain amount of RAM, cpu
ram
filestorage
All of the above things are part of the deployment
K8s will keep track of it
Deployment is something that keeps on going.
It has a Deployment controller, that in essense
If a container goes down, K8s will know about it will try to do anything to auto heal it
It iwll spin an another container up and will try to auto heal it.
So Deployment is not something of initial launch of a container, but something that keeps on going.
Scaling:
--------
If we are getting more traffic and we need to scale up.
Sometimes we are going to be in a situation where, we have 1 server per container
Its not cost effective.
Scaling a deployment is done by modifying a deployment.
It also can hold on to the scale.
How many CPUs it needs, how much RAM it needs.
I need to deploy 5 different nginx containers, you put them in the best spot, given the following requirements.
How am i going to connect to a container.
Nginx is one of our services.
We have to connect to them
Services let us manage all of these.
We can have multiple services, we can have multiple deployments.
W
We have multiple nodes that are running.
Scaling a Dep
Taking the experiences from the past and predicting the future is called Decision Making
It can work with any type of container, not only docker
At its core, its gives you a platform that can
1) Scale easily
2) Easy to do monitoring
3) Means to do deployments
Master node is the kubernetes Master
You decide how to do stuff
You talk to Kubernetes and tell it what kind of image you want to use to create a container from and give it other critereon
It creates a deployment
Deployment can say a bunch of different things
i need a certain amount of RAM, cpu
ram
filestorage
All of the above things are part of the deployment
K8s will keep track of it
Deployment is something that keeps on going.
It has a Deployment controller, that in essense
If a container goes down, K8s will know about it will try to do anything to auto heal it
It iwll spin an another container up and will try to auto heal it.
So Deployment is not something of initial launch of a container, but something that keeps on going.
Scaling:
--------
If we are getting more traffic and we need to scale up.
Sometimes we are going to be in a situation where, we have 1 server per container
Its not cost effective.
Scaling a deployment is done by modifying a deployment.
It also can hold on to the scale.
How many CPUs it needs, how much RAM it needs.
I need to deploy 5 different nginx containers, you put them in the best spot, given the following requirements.
How am i going to connect to a container.
Nginx is one of our services.
We have to connect to them
Services let us manage all of these.
We can have multiple services, we can have multiple deployments.
W
We have multiple nodes that are running.
Scaling a Dep
Taking the experiences from the past and predicting the future is called Decision Making
Comments
Post a Comment