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
We can statically analyze the image entrypoint dependency, or use runtime dynamic analysis technologies such as ebpf, fanotify, metric, etc. to obtain the container file access pattern, and then build this part of data into an independent image layer:
preferentially fetch blob during the image startup phase to reduce network and disk IO.
avoid frequent image builds, allows for better local cache utilization.
Possible implementation steps:
follow the doc to get an accessed files list for a OCI / nydus image.
implement cat accessed-files.txt | nydusify optimize --policy separated-prefetch-blob --target $existed-nydus-image --output $new-nydus-image to generate a new image, which references a new blob included prefetch file chunks.
The text was updated successfully, but these errors were encountered:
imeoer
changed the title
[OSPP 2024] improve image build for prefetch
[OSPP 2024] optimize image by prefetch analysis
Jun 18, 2024
We can statically analyze the image entrypoint dependency, or use runtime dynamic analysis technologies such as ebpf, fanotify, metric, etc. to obtain the container file access pattern, and then build this part of data into an independent image layer:
Possible implementation steps:
cat accessed-files.txt | nydusify optimize --policy separated-prefetch-blob --target $existed-nydus-image --output $new-nydus-image
to generate a new image, which references a new blob included prefetch file chunks.The text was updated successfully, but these errors were encountered: