Re: CDR with many tables
Madison Pruet said an upcoming release would let templates define replication for many tables on many nodes with two commands and online sync. Malc_p said he really wanted experience of CDR with many tables under OLTP. Madison knew of users with many replicates but could not name them and listed complications of multi-table setup. Malc_p said his master-slave setup would be initialised from a level-0 backup restore.
Auto-generated by Claude from the posts below — may be imperfect; read the full thread.
In a near upcoming release, you can utilize templates to define replication for 'x' number of tables across 'x' number of nodes using only 2 commands. Not only that, but the two commands can be used to perform an online sync of the table. M.P. <malc_p@btinternet.com> wrote in message news:1108389507.349208.316000@g14g2000cwa.googlegroups.com... > IDS9.30HC5; HP-UX11i > This is actually a follow-up to my previous regarding > deferred level-1 restores, and is associated with the > resultant "HDR not allowing blobspaces" threads. > I know CDR allows blobs and I know it works because > I've done some testing previously. > But...... CDR requires a separate definition for every > table in every database you want to replicate. I have > scripted the process and that is also no problem. > What I wondered was, how would CDR cope with upwards > of 650 tables defined in an OLTP system? We can get > through 50 x 10Mb logs an hour at peak times.... > Network capacity isn't a problem though. > Anyone doing CDR with lots of tables? > Cheers > Malc >
Yup fine but that's not what I was asking. It's a trivial task, really, to script some templates. I just wondered what advice/experiences people had on using CDR with (what seems to me to be) a large number of tables running OLTP.
ok - understood. I know of folks using ER with a larger number of replicates, but am not able to give names (legal reasons) The task of creating multi-table replicates accross multi-nodes is a bit more complicated than you might think, however. ;-) You have to 1) make sure the column types match up, including constraint rules 2) optionally create the tables 3) perform an initial sync on the set of tables in such a way that referential integrety rules are supported. 4) if the tables on the vairous nodes already contain data, determine how you will deal with extra rows on the various nodes (i.e. eliminate, merge, ignore...) 5) If you have to get rid of extra rows on the nodes, then how do you deal with dependent tables which might contain children rows -- even if cascading deletes aren't turned on 6) as you are copying the rows from one server to another during the sync, how do you deal with triggers? ... and the list goes on... ;-) <malc_p@btinternet.com> wrote in message news:1108457154.053569.278590@o13g2000cwo.googlegroups.com... > Yup fine but that's not what I was asking. It's a trivial task, really, > to script some templates. > I just wondered what advice/experiences people had on using CDR with > (what seems to me to be) a large number of tables running OLTP. >
Luckily (for me as DBA!) it's just a master-slave setup and we'll be initializing it with a level-0 backup/ imported restore process over a nice quiet weekend with none of those awkward users getting in the way.....
lucky you.... ;-) <malc_p@btinternet.com> wrote in message news:1108485471.353008.279620@o13g2000cwo.googlegroups.com... > Luckily (for me as DBA!) it's just a master-slave setup and we'll be > initializing it with a level-0 backup/ imported restore process over a > nice quiet weekend with none of those awkward users getting in the > way..... >