Kubernetes Course: Difference between revisions
Adamcolton (talk | contribs) Created page with "https://www.coursera.org/learn/google-kubernetes-engine == Intro To Containers == Dedicated servers can be have inconsistently, so we solved that with Virtual Machines. But o..." |
Adamcolton (talk | contribs) No edit summary |
||
| (One intermediate revision by the same user not shown) | |||
| Line 5: | Line 5: | ||
== Intro to Docker == | == Intro to Docker == | ||
Docker files. In practice, push and pull images from registry. Containers are efficient and portable. | |||
could not do lab | |||
== Clusters, Nodes and Pods == | |||
The master controls the kublets. A cluster is a set of computers (pods) with a master. A cluster can have multiple masters. Can be thousands of nodes. It can allocate jobs that are high on resources or remove jobs from nodes that are low on resources. | |||
K8n uses yaml files. | |||
== Model == | |||
(not a lecture, but I'm putting it here | |||
A cluster contains nodes. A node contains a kubelet and many pods. Pods contain applications (I think just one). A cluster can contain networking services and data storage services. | |||
== Services, Labels and selectors == | |||
A pod has a fixed IP. Labels are key/value meta-data for grouping, filtering and searching. | |||
== Volumes == | |||
Docker provides data storage for containers. K8n volumes allow containers to share data. Volumes are attached to pods before they are brought online. | |||
[[Category:Course Work]] | [[Category:Course Work]] | ||
Latest revision as of 19:49, 26 June 2018
https://www.coursera.org/learn/google-kubernetes-engine
Intro To Containers
Dedicated servers can be have inconsistently, so we solved that with Virtual Machines. But one server running many virtual machines will have a lot of redundancies. A container environment can share those redundancies (primarily the kernal and container runtime).
Intro to Docker
Docker files. In practice, push and pull images from registry. Containers are efficient and portable.
could not do lab
Clusters, Nodes and Pods
The master controls the kublets. A cluster is a set of computers (pods) with a master. A cluster can have multiple masters. Can be thousands of nodes. It can allocate jobs that are high on resources or remove jobs from nodes that are low on resources.
K8n uses yaml files.
Model
(not a lecture, but I'm putting it here
A cluster contains nodes. A node contains a kubelet and many pods. Pods contain applications (I think just one). A cluster can contain networking services and data storage services.
Services, Labels and selectors
A pod has a fixed IP. Labels are key/value meta-data for grouping, filtering and searching.
Volumes
Docker provides data storage for containers. K8n volumes allow containers to share data. Volumes are attached to pods before they are brought online.