2006/3/22, "Döhr, Markus ICC-H" <[EMAIL PROTECTED]>: > > > MaxDB has a complete different data storage approach as e. > > g. Oracle or DB2. > > > There are not tablespaces, datafiles, extents, and > > controlfiles. MaxDB > > > uses Volumes and spreads the data accross those volumes. > > You have no > > > influence where a table is stored so you will never run > > into "hotspots". > > > > But it also means that you have no chance to optimize it if > > the usage pattern is bad; and unfortunately there is no way > > to change the size of a volume (is there?). I personally > > found SQL Server to be the most hassle free RDBMS I have > > worked with. I'm sorry to say that on this list but also YMMV... > > What do you mean with "usage pattern"?
More inserts or queries, which tables etc. > Usually (in 99 % of the cases) this > is not necessary since the I/O is spread always across all volumes. A > (parallel) access to read from N volumes is much faster than a single read > on a single disk. Only if the IO subsystem provides the bandwidth. This is also only true if those volumes reside on different disks, if you have multiple volumes on the same disk it's likely to be slower. For example, if you create a volume and detect later that the initial size was too small, what do you do? I know you can create a new, larger volume with 7.6 and drop the old one but this has quite some impact with regard to IO and space. Is there a way to resize a volume? I couldn't find one yet. Also, you might want to reserve bandwidth for certain table accesses by placing them on separate volumes / table spaces / file groups. You cannot do that with MaxDB, can you? > How often do you need to move tables from one tablespace to another - in > real life? The frequency of the question in several DB related newsgroups I read seems to indicate that it's a fairly common problem. Kind regards robert -- Have a look: http://www.flickr.com/photos/fussel-foto/ -- MaxDB Discussion Mailing List For list archives: http://lists.mysql.com/maxdb To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
