Kubernetes Master Class - Use persistent storage In Kubernetes and Project Longhorn.pptx

jvntecnologia 63 views 19 slides Jul 02, 2024
Slide 1
Slide 1 of 19
Slide 1
1
Slide 2
2
Slide 3
3
Slide 4
4
Slide 5
5
Slide 6
6
Slide 7
7
Slide 8
8
Slide 9
9
Slide 10
10
Slide 11
11
Slide 12
12
Slide 13
13
Slide 14
14
Slide 15
15
Slide 16
16
Slide 17
17
Slide 18
18
Slide 19
19

About This Presentation

Kubernetes Master Class - Use persistent storage In Kubernetes and Project Longhorn.pptx


Slide Content

July 16, 2019 #Rancherk8s Use persistent storage In Kubernetes and Project Longhorn Kubernetes Master Class Series

Matthew Scheer Marketing Manager     @ matthew_scheer [email protected] Sheng Yang Software Architect / / : @ yasker #RancherK8s

Rancher Master Class Series: Trying to keep this to 40-45 minutes Questions are always welcome Use the questions tab to write your questions We may respond to all, so mark your question as private if needed. #Rancherk8s

This session is being recorded! http://youtube.com/c/rancher #Rancherk8s

Join the conversation on Slack http://slack.rancher.io #masterclass #RancherK8s

Upcoming Classes http://rancher.com/kubernetes-master-class/ #RancherK8s

Agenda Persistent storage concepts in Kubernetes RWO vs RWX, Deployment vs StatefulSet Distributed Block Storage Software: Longhorn #RancherK8s

Persistent storage in Kubernetes Persistent Volume Persistent Volume Claim Storage Class Provisioner Volume

Persistent storage in Kubernetes before StorageClass PVC PV PVC PV PVC PV PVC PV POD Volume Volume POD Volume POD Volume Admin

Persistent storage in Kubernetes after StorageClass PVC PV PVC PV PVC PV PVC PV POD Volume Volume POD Volume POD Volume Storage Class with Provisioner

Read-Write-Once (RWO) vs Read-Write-Many (RWX) Read-Write-Once type storage only can be read/write on one node at any given time High performance block device AWS EBS, Azure Disk, Google Persistent Disk, Ceph RBD, Longhorn Read-Write-Many type storage can be read/write on multiple nodes at the same time Distributed Filesystem AWS EFS, NFS, GlusterFS , CephFS

Deployment vs StatefulSet Pods in one Deployment share the same volume No matter which node the pod runs on Better suit for RWX type storage Each Pod in one StatefulSet can have one volume VolumeClaimTemplate Better suit for RWO type storage

Open Source Distributed Block Storage Software For Kubernetes https://github.com/rancher/longhorn/ Add persistent storage support to any Kubernetes cluster kubectl apply –f longhorn.yaml

Latest release: Longhorn v0.5.0 Enterprise-grade distributed block storage software for Kubernetes Volume snapshots Volume backup and restore Live upgrade of Longhorn software without impacting running volumes Cross-cluster disaster recovery volume with defined RTO and RPO Intuitive UI One click installation And more features are coming QoS, volume resizing, real time performance monitoring, et cetera.

Longhorn Architecture - Engine Pod 3 Engine Replica Replica Pod 1 Engine Replica Replica Pod 2 Engine Replica Replica Volume Volume Volume Node 1 Node 2 Orchestrated by Kubernetes

Longhorn Architecture - Manager Longhorn UI Kubernetes Cluster Longhorn Manager (Orchestrates all the volumes) Longhorn API Longhorn API Engine Replica 1 Replica 2 Engine Replica 1 Replica 2 Engine Replica 1 Replica 2 Kubernetes API Server … Longhorn CSI Plugin Container Storage Interface API

Demo

Upcoming Longhorn v0.6.0 (Beta) Re-architecture Engines and replicas would be run as processes inside the DaemonSet Pods Instead of one pod for each engine or replica Result Speed up volume attach/detach process No more worry about Pod per node limitation Guaranteed resource for DaemonSet Pods Storage Pool support

Thank you @ Rancher_Labs   ·   #Rancherk8s Rancher.com/ kubernetes -master-class
Tags