comp.databases.kerry.sainsbury and tbtape question

[...] > If the data base is fragmented as it sits physically on the disk, does > the tbtape PROCESS unfragment the data as it is streamed to the > tape? > > Or is it streamed block by block in its fragmented condition? > > Would streaming it back to the disk if it WAS unfragmented to the > tape be refragmented if the extents were not big enough? > > Why would I ask? Cause maybe I'm gonna reformat the hard drive(s) > and restore the data base and wonder if maybe the data will be > restored in an UNfragmented form even if the extents are not up to > par. ( Yes maybe I should recreate the db with new extents ... :- ) > > Happy Holidays! > > timmy > The Computer Business Company > > Any reindeer coming close to my house will probably sweat.... The way I understand it an archive will NOT cure disk fragmentation. There are a couple of different kinds of fragmentation, though. There is the fragmentation of rows within a contiguous space (usually cured by altering the table: sql ALTER TABLE, OL Admin Guide v5.0 pg 3-68) and the fragmentation of extents within a dbspace (cured by unloading, dropping, re-creating, and re-loading the table.) You should set up your table to fit within (at best) a single, congiguous extent. You can size the initial extent at table creation time. There is a penalty if your table has more than 8 extents. This has been a gross simplification of the whole problem and solution, so use with care. Good luck, __________________________________________________________________ | Clem Akins Standard Disclaimers Apply | |Reynolds Metals Co, Alloys Plant "Climb High, Cave Deep!" | | Muscle Shoals, Alabama USA cwakins@leia.alloys.rmc.com | |________________________________________________________________|