The idea is to not give control to the user which cloud his data is going to. That detail is useless for the user particularly if we are claiming to manage data storage based on network conditions.
cf add cf remove cf add -i cf add --imp
List that flat file
The command will be implemented such that it will go to the cloud and list the buckets intensionally. Do not display from a local copy.
This should show the entire namespace cf ls Remove dups
This should list the entire namespace segragated by which cloud profile has what This will be useful for the demo cf ls -s cf ls --seg Show all files in all clouds
IP addrs - Endpoint Ping time -------- inconsistent Upload download ----- Client server model for CLI Server contacts cloud only when CLI reqs Server periodically updates from list of files
CLI runs then exits Who will our CLI ask the network stats from: A running process - Who starts it? What happens if it crashed? Read from a file that a background process keeps updating - Synchronization
Fetch file from cloud Fetch imp from nearest cloud
Automatic background running server which updates files on the cloud or resurects them