The Register® — Biting the hand that feeds IT

Feeds

Sun setting dedupe up for ZFS

Oracle/Sun's ZFS file system seems set to get deduplication added to it later this year. ZFS or the Zettabyte File System is a 128-bit file system that Sun says radically simplifies file system administration. Amongst its features are 256-bit checksums that detect and correct silent data corruption. In other words, it is already …

This topic is closed for new posts.
Anonymous Coward
Thumb Up

Yay!

Now all we need is a good open-source storage oriented web interface / gui to be created and we can build our own OpenStorage (well close enough) systems at a much much lower price.

Lou 2
Big Brother

If it was that easy ...

Hmmm, actually it is - a few 1TB disks from the corner electronic store, a few Linux clones and I have my own backup Saas service - watch-out Google and cronies.

Now lets see if Oracle recognizes that with RAC, OpenStorage, their DB files, applications and log files spread with some ZFS goodness and they can take over the world. Bet $5 they won't!

Thinking off it - they already own have the world anyway.

Anonymous Coward
Stop

Re: Yay!

And you'll get the analytics from where???

There's absolutely nothing stopping you from taking the same building blocks that the FISHworks team did and building your own appliance but the big piece that would be missing is the performance analytics which are really the USP in the Unified Storage System.

Good luck on doing that from scratch!

David Halko

ZFS with a good storage oriented GUI...

ZFS with a good storage oriented GUI already exists, in the Sun Open Storage line.

With features like snapshot & cloning already solid - dedup is not much more sophisticated.

Dedup with Zones, LDom's, and xVM is a huge bonus! With Virtual Box or Thin Clients like the Sun Ray - the benefits are astounding to businesses!

Run 100 virtual servers or 100 virtual desktops on a server, served by a de-duped ZFS filesystem, and the common binaries would easily fit in a very small memory footprint since only one copy of each commonly used file would need to be in memory.

Dedup with clustering would be absolutely phenomenal!!! (We have to wait a little while longer for this one until Lustre source code is sufficiently leveraged...)

This topic is closed for new posts.