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
At present, our filesystem mount process is:
step 1. pull CloudImage and mount all layers.
step 2. then send mounted data as rootfs, and execute the init script to start up cri, kubelet and so on.
step 3. send registry data and will be loaded at runtime ApplyRegistry to start registry.
we can try to optimize above process i think.
一:for step2, Sort rootfs by the file usage:
cluster node initialization file:like CRI and kubelet configs ,scripts and binary files. send to each cluster nodes.
user application startup file: user application startup data used at the guest module.like manifests,charts. those file no need to send.
二:for step3, strip registry data send from filesystem send.
send registry data and apply start registry at the same process reducing the complexity of filesystem module.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
background:
At present, our filesystem mount process is:
step 1. pull CloudImage and mount all layers.
step 2. then send mounted data as rootfs, and execute the init script to start up cri, kubelet and so on.
step 3. send registry data and will be loaded at runtime
ApplyRegistry
to start registry.we can try to optimize above process i think.
一:for step2, Sort rootfs by the file usage:
二:for step3, strip registry data send from filesystem send.
send registry data and apply start registry at the same process reducing the complexity of filesystem module.
Beta Was this translation helpful? Give feedback.
All reactions