Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

What ever happened to versioning file systems?

shovas (1605685) writes | more than 4 years ago

Data Storage 3

shovas (1605685) writes "As a developer and sysadmin, the benefits of revision control systems are clear. It only seems natural that a simpler, transparent approach to versioning files on a regular file system would be a net win. There's ext3cow and Wayback FS, and possibly some fuse-based projects, but each is either dead, immature or just not applicable. So what happened to the promise of versioning file systems? Hasn't everyone lost a file to a bad rm command? Hasn't everyone wished they could see a revision of a file in the past? What's the hold up?"

Sorry! There are no comments related to the filter you selected.

Competition from "snapshotting" filesystems (1)

Bandraginus (901166) | more than 4 years ago | (#31069284)

It really seems as though the market has moved more towards snapshotting file systems (think ZFS, etc). I guess it's easier to deal with (technically) at the block level, plus easier for end users to understand and use (ever tried to get the hoi polloi to use a decent revision control system?). The end result is nearly the same for most practical applications, and ends up being faster (I have no citation to give you, but the fact that there *is* no mature versioned file system is my evidence. Must be hard to write an efficient diffing algorithm that works for most file types).

Re:Competition from "snapshotting" filesystems (1)

AliasMarlowe (1042386) | more than 4 years ago | (#31070238)

Must be hard to write an efficient diffing algorithm that works for most file types

What has a diff algorithm got to do with a versioning file system? Every time a file is saved, it should be written as a new version. Diff would be used to compare two versions of a file, not to decide whether to make a new version.

An ancient approach was used in DEC's FILES-11 system back in the 80's, where each file name had a version number attached (such as FILE.EXT;158). Files could be specified explicitly including the version number, but if a file was specified without a version number, then the highest version was implied. A simple purge command would delete older versions keeping a specified number of the most recent versions (in RSX-11, the command was PIP FILE.EXT/PU:5 to delete all but the 5 highest versions). There was also an open-in-place feature so that special applications - databases and the like - could repeatedly modify large files without making a new version.

Really, I'd like to see versioning file systems again.

Re:Competition from "snapshotting" filesystems (1)

uassholes (1179143) | more than 4 years ago | (#31129832)

File versioning did not die with RSX-11. It continued with DEC's (now HP) RMS file system used in their OpenVMS OS, which is still very much in use on DEC Alpha and Intel Itanium CPUs. []
Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?