Linked by Eugenia Loli-Queru on Thu 15th Sep 2005 21:14 UTC
A lot of hooplah has been distributed by Sun Microsystems on the advantages of their OpenSolaris/Solaris 10 release. Martin C. Brown has been using said software for the past few months and files his review that helps answer the question: has someone finally found a Linux killer?
Thread beginning with comment 32045
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
Recent Original Stories
Recent Comments
Headlines
Random Comments
Random Stories
Random OS Link
Member since:
2005-07-08
"First off Sun is all about PR, they hyped up zfs when it was basically vap-o-ware."
For filesystem design, the devil is in the details. I'm sure that ZFS looked like 12 person-years on paper, and that's what management based their marketing on. However, in reality it is 12 py to design and implement the filesystem and 18 py to figure out why it isn't performing as well as predicted. You gotta get the on-disk format nailed down, the kernelspace can be released as reference and tuned through beta testing. But once you get the reference kernelspace, there's inevitably a point at which you realize the shortcomings of the on-disk format. It takes two iterations to get it right, and then you get to worry about the userspace commands...
Sun just realized the hard way that filesystems are tough. When you look back on the ZFS hype, there was a lot of hype all of sudden (they were close to a working kernelspace), and then there was the admission that it would be late (back to the drawing board). The hype died down to a trickle of interview responses declaring that ZFS is not dead, it's just taking longer than expected. Microsoft did the same thing with WinFS, and it took 10 years to get from Reiser3 to Reiser4. Welcome to the world of filesystems.