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
When using the app on a cluster I occasionally get a disk I/O error (screenshot attached), suggesting an issue with the disk caching we're using to store the computed dataframes on NFS.
The text was updated successfully, but these errors were encountered:
My 10c here is that the SQL based approach is adding some unnecessary complexity, with file locks and so on (which NFS doesn't like). A simple file-based approach would be simpler and more debuggable.
When using the app on a cluster I occasionally get a disk I/O error (screenshot attached), suggesting an issue with the disk caching we're using to store the computed dataframes on NFS.
The text was updated successfully, but these errors were encountered: