You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
What would you like to be added:
It is difficult for users to operate, manage and upgrade the K8s cluster in their own IDC (Internet Data Center). We need to enhance the operational efficiency of K8s cluster in user's IDC by adopting KOK (Kubernetes-On-Kubernetes) architecture, where there are host-K8s and tenant-K8s. Host-K8s is located at cloud, provided by cloud service providers and can manage control plane of tenant-K8s in IDC.
Why is this needed:
For K8s clutesrs in user's IDC, it is difficult to operate, manage and upgrade the control plane components. Users typically adopt the following three solutions to manage K8s clusters in their IDC.
Some users only set up a single K8s cluster in IDC for tenant. In this case, when K8s have version upgrades and changes, about three major releases per year, users will suffer from complex operations to upgrade those components. Meanwhile, there is no resource elasticity capability in K8s clutesrs in user's IDC, such as scaling control plane components, which is a costly operation for user.
Some users adopt the KOK architecture in their own IDC to manages tenant-K8s's control plane components. Both host-K8s and tenant-K8s are in user's IDC. In this case, operating and updating control plane components of tenant-K8s will be easy, however, it is still hard to operate and upgrade the control plane components in host-K8s.
More and more users only access their IDC machines to cloud service providers as worker nodes, utilizing the abilities of cloud-edge collaboration provided by OpenYurt. But there are some users needs continuous deployment for offline tasks, depending on strong stability of cloud-edge communication, in this case, they tend to maintain a K8s cluster in their IDC.
We aim to solve the pain points mentioned above, which automates the operation and maintenance of control plane components of tenant-K8s to replace manual user operations, and affords users who needs continuous deployment for offline tasks a efficient operation scheme to manage their IDC K8s cluster.
others
/kind feature
The text was updated successfully, but these errors were encountered:
What would you like to be added:
It is difficult for users to operate, manage and upgrade the K8s cluster in their own IDC (Internet Data Center). We need to enhance the operational efficiency of K8s cluster in user's IDC by adopting KOK (Kubernetes-On-Kubernetes) architecture, where there are host-K8s and tenant-K8s. Host-K8s is located at cloud, provided by cloud service providers and can manage control plane of tenant-K8s in IDC.
Why is this needed:
For K8s clutesrs in user's IDC, it is difficult to operate, manage and upgrade the control plane components. Users typically adopt the following three solutions to manage K8s clusters in their IDC.
Some users only set up a single K8s cluster in IDC for tenant. In this case, when K8s have version upgrades and changes, about three major releases per year, users will suffer from complex operations to upgrade those components. Meanwhile, there is no resource elasticity capability in K8s clutesrs in user's IDC, such as scaling control plane components, which is a costly operation for user.
Some users adopt the KOK architecture in their own IDC to manages tenant-K8s's control plane components. Both host-K8s and tenant-K8s are in user's IDC. In this case, operating and updating control plane components of tenant-K8s will be easy, however, it is still hard to operate and upgrade the control plane components in host-K8s.
More and more users only access their IDC machines to cloud service providers as worker nodes, utilizing the abilities of cloud-edge collaboration provided by OpenYurt. But there are some users needs continuous deployment for offline tasks, depending on strong stability of cloud-edge communication, in this case, they tend to maintain a K8s cluster in their IDC.
We aim to solve the pain points mentioned above, which automates the operation and maintenance of control plane components of tenant-K8s to replace manual user operations, and affords users who needs continuous deployment for offline tasks a efficient operation scheme to manage their IDC K8s cluster.
others
/kind feature
The text was updated successfully, but these errors were encountered: