Merge pull request #192 from llhuii/allinone-script

Add all-in-one installation script/docs

README.md

@@ -78,7 +78,7 @@ Documentation is located on [readthedoc.io](https://sedna.readthedocs.io/). Thes
 
 
 ### Installation
-Follow the [Sedna installation document](docs/setup/install.md) to install Sedna.
+Follow the [Sedna installation document](docs/setup/quick-start.md) to install Sedna.
 
 ### Examples
 Example1：[Using Joint Inference Service in Helmet Detection Scenario](/examples/joint_inference/helmet_detection_inference/README.md).  

docs/setup/all-in-one.md

@@ -0,0 +1,38 @@
+### Deploy All In One Sedna
+The [all-in-one script](/scripts/installation/all-in-one.sh) is used to install Sedna along with a mini Kubernetes environment locally, including:
+  - A Kubernetes v1.21 cluster with multi worker nodes, default zero worker node.
+  - KubeEdge with multi edge nodes, default is latest KubeEdge and one edge node.
+  - Sedna, default is the latest version.
+
+It requires you:
+  - 2 CPUs or more
+  - 2GB+ free memory, depends on node number setting
+  - 10GB+ free disk space
+  - Internet connection(docker hub, github etc.)
+  - Linux platform, such as ubuntu/centos
+  - Docker 17.06+
+
+For example: 
+
+  ```bash
+  curl https://raw.githubusercontent.com/kubeedge/sedna/master/scripts/installation/all-in-one.sh | NUM_EDGE_NODES=2 bash -
+  ```
+
+Above command installs a mini Sedna environment, including:
+  - A Kubernetes v1.21 cluster with only one master node.
+  - KubeEdge with two edge nodes.
+  - The latest Sedna.
+
+You can play it online on [katacoda](https://www.katacoda.com/kubeedge-sedna/scenarios/all-in-one).
+
+Advanced options:
+| Env Variable |  Description| Default Value|
+| --- |  --- | --- |
+|NUM_EDGE_NODES     | Number of KubeEdge nodes| 1 |
+|NUM_CLOUD_WORKER_NODES    | Number of cloud **worker** nodes, not master node| 0|
+|SEDNA_VERSION    | The Sedna version to be installed. |The latest version|
+|KUBEEDGE_VERSION    | The KubeEdge version to be installed. |The latest version|
+|CLUSTER_NAME       | The all-in-one cluster name| sedna-mini|
+|FORCE_INSTALL_SEDNA       | If 'true', force to reinstall Sedna|false|
+|NODE_IMAGE       | Custom node image| kubeedge/sedna-allinone-node:v1.21.1|
+|REUSE_EDGE_CONTAINER      | Whether reuse edge node containers or not|true|

docs/setup/install.md

@@ -1,4 +1,10 @@
-This guide covers how to install Sedna on existing KubeEdge environment, i.e. its two main components: [GM(GlobalManager)](/README.md#globalmanager) and [LC(LocalController)](/README.md#localcontroller).
+This guide covers how to install Sedna on an existing Kubernetes environment.
+
+For interested readers, Sedna also has two important components that would be mentioned below, i.e., [GM(GlobalManager)](/README.md#globalmanager) and [LC(LocalController)](/README.md#localcontroller) for workerload generation and maintenance.
+
+If you don't have an existing Kubernetes, you can:
+1) Install Kubernetes by following the [Kubernetes website](https://kubernetes.io/docs/setup/).
+2) Or follow [quick start](quick-start.md) for other options.
 
 ### Prerequisites
 - [Kubectl][kubectl] with right kubeconfig

docs/setup/local-up.md

@@ -1,26 +1,48 @@
-### Deploy local Sedna clusters
+### Deploy Local Sedna Cluster
 
-Once you have docker running, you can create a local Sedna cluster with:
+The [local-up script](/hack/local-up.sh) boots a local Kubernetes cluster, installs latest KubeEdge, and deploys Sedna based on the Sedna local repository.
+
+#### Use Case
+When one is contributing new features for Sedna, codes like AI algorithms under testing can be frequently changed before final deployment.
+When coding in that case, s/he would suffer from tortured re-installations and frequent failures of the whole complicated system.
+To get rid of the torments, one can use the local-up installation, embraced the single-machine simulation for agiler development and testing.
+
+#### Setup
+
+It requires:
+  - 2 CPUs or more
+  - 1GB+ free memory
+  - 5GB+ free disk space
+  - Internet connection(docker hub, github etc.)
+  - Linux platform, such as ubuntu/centos
+  - Docker 17.06+
+  - A local Sedna code repository
+
+
+Then you can enter Sedna local code repository, and create a local Sedna cluster with:
 ```
 bash hack/local-up.sh
 ```
-This script uses [kind](https://kind.sigs.k8s.io/docs/user/quick-start/) to create a 
-local k8s cluster with one master node, and boots one edge node by running KubeEdge.
-You can see them by using `kubectl get nodes -o wide`:
+
+In more details, this local-up script uses [kind](https://kind.sigs.k8s.io/docs/user/quick-start/) to create a 
+local K8S cluster with one master node, and joins the K8S cluster by running KubeEdge.
+
+In another terminal, you can see them by using `kubectl get nodes -o wide`:
 ```shell
 NAME                  STATUS   ROLES                  AGE     VERSION                   INTERNAL-IP     EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION       CONTAINER-RUNTIME
 edge-node             Ready    agent,edge             3d21h   v1.19.3-kubeedge-v1.6.1   192.168.0.233   <none>        Ubuntu 18.04.5 LTS   4.15.0-128-generic   docker://20.10.2
 sedna-control-plane   Ready    control-plane,master   3d21h   v1.20.2                   172.18.0.2      <none>        Ubuntu 20.10         4.15.0-128-generic   containerd://1.5.0-beta.3-24-g95513021e
 ```
 
-You can access master node with:
+You can login the master node with:
 ```
 docker exec -it --detach-keys=ctrl-@ sedna-control-plane bash
+# since the master node just uses containerd CRI runtime, you can alias the CRI cli 'crictl' as 'docker'
 alias docker=crictl
 ```
 
-Docker images can be loaded into the cluster nodes with:
+After you have done developing, built worker image and want to run your worker into master node, your worker image should be loaded into the cluster nodes with:
 ```
-kind load docker-image my-custom-image --name sedna
+kind load docker-image --name sedna <your-custom-worker-image>
 ```
 

docs/setup/quick-start.md

@@ -0,0 +1,10 @@
+There are some ways to set up Sedna, depends on your use case:
+- If you have none Kubernetes environment and don't want to install Kubernetes manually, you can follow [the instruction](./all-in-one.md) to install all-in-one Sedna environment.
+
+- Otherwise, when Kubernetes is ready, you can follow [this instruction](./install.md) to install Sedna on the existing Kubernetes cluster.
+
+- One more thing: we also have a local-up install script available for easier Sedna development. The local-up version boots a local Kubernetes cluster and installs Sedna based on a local repository. The use case and setup procedures are available [here](./local-up.md).
+
+
+After set up Sedna, you can try out [examples](/examples).
+