On 03/22/12 13:13, Dmitry Yemanov wrote: > 22.03.2012 13:00, Thomas Steinmaurer wrote: > >> * Under the assumption that SS as a single-process model with it's >> shared cache supports SMP (I think this was a primarly goal) in 3.0 like >> SC in 2.5, I don't see a reason to offer SC in 3.0 at all (at least the >> SC terminology; which already shows up in the missing SC part in "instsvc")? > Personally, I agree. But we need to hear other opinions as well. >
There may exist various configurations. Specially with per-database selectable mode of operation. For example, we have 2 databases, one big and used in SS mode, another relatively small, but it must be accessed by WWW server. The fastest way for it is embedded access, but sometimes we need to update information in that database too (typically pass something from main database). Making it superclassic can be ideal in this case. We anyway have single process executing all network requests which works in shared cache (SS) mode with first database, but can also access second database as classic one. Page cache for main process (WWW), accessing second database, can be done big enough. ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
