Actually, the IBM (and related) mainframe file systems were built on the same basis as their earlier (non-RDBMSs) databases, and their VM/ESA SFS and BFS (Shared File System and Byte-oriented File System respectiviely) are clients/front-ends to a backend based on the SQL/DS RDBMS.
Microsoft here is very much the Johnny-come-lately. Wesley Parish On Fri, 31 Oct 2003 15:02, you wrote: > On Fri, Oct 31, 2003 at 02:27:25PM +0000, Jason Greenwood wrote: > > It covers MS plans for Longhorn and it's > > 'kernel level' inbuilt search ability. As if the NT kernel was not > > bloated enough already... > > that's skewing it abit. > the kernel level search ability is related to the fact that the > filesystem has more database capabilities. > > finding a file on disk is what a filesystem is about. > linux has filesystems. > linux finds files. > linux has a kernel level search ability. > (or where do you think does the path to a file get interpreted?) > > this is not a question if the kernel has search abilites, but only how > good they are. the better they are, the more flexible you can lay out > your data. > > the idea of a database like fs is not new, BeOS already had it. > > and i'll take a filesystem that supports files with arbbitrary metadata > which is efficiently searchable any day over todays offerings. > (now of course with linux userland fs capabilities it is probably not > necessary to build support for such a database fs into the kernel, but > would you install linux on a filesystem that is not supported by linux > natively? i am not so sure) > > greetings, martin. -- Clinesterton Beademung - in all of love. Mau e ki, "He aha te mea nui?" You ask, "What is the most important thing?" Maku e ki, "He tangata, he tangata, he tangata." I reply, "It is people, it is people, it is people."
