This is a project for deploying a development for AI, especially for the case of internal network.
So, generally, the environment is as following:
- One machine for deploying docker registry with the pull cache mode. Refer to registry .
- One machine for installing Rancher. Refer to rancher .
- (Optional) If needed, make the ssh tunnel for accessing the Rancher. Refer to sshtunnel .
Plan:
- (1) Docker Registry (in proxy machine) .
- (2) Nexus3-helm-repo (in proxy machine, as the proxy of Docker, Helm, Yum, Maven, etc.)
- (3) Rancher (in production machine, management of the clusters of Kubernetes)
- (4) chartmuseum (in production machine, a tools for the installation of helms)
- (5) Openproject (for project management)
- (6) Eclipse-Che (Web IDE)
- (7) Kong (API gateway, service mesh)
- (8) Istio (Service Mesh)
- (9) Ceph (Storage，怎么安装，kubernetes?)
- (10) Openshift ()
- (11) GitLab ()
- (12) Bitbucket ()
- (13) Jenkins ()
- (14) codis ()
- (15) Ambari ()
- (16) Gogs (Git Server written by GO)
- (17) Jupyterhub ()
- (18) Jupyterlab ()