> Just my opinion, of course. I'd be interested to hear how much effect > this would have on others and whether they think it's worthwhile to even > investigate.
Moved and seconded. I don't think the world needs to reinvent packed-decimal yet again and all the grief that fancy encoding schemes cause, especially since they tend to fail miserably on different-endian machines (see ext4 problems on non-Intel architectures -- you were planning on testing this stuff on PPC and zSeries, right?). This is a database tuning issue, really -- if your database isn't set up to handle the volume Bacula generates, then it's time. IMHO, you'd get more bang for the work by abstracting the file parameters into a list of attributes and pointing to those via a table. That would also give you a lot of the infrastructure necessary to handle the arbitrary attribute support you need for SELinux, etc, and give you a good start on being able to deduplicate file data (no point in storing the same file multiple times). ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Bacula-devel mailing list Bacula-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-devel
