Performance for Transactional replication for indexed view
in [Replication]
|
Prev: SQL 2005 subscriber and SQL 2000 publisher/distributor - Could not connect to distributor.
Next: 'INITCOLVS' is not a recognized built-in function name
From: mecn on 19 Sep 2006 09:57 Hi all, We are planning to have transactional repllication for 8 tables(indexed-view rep) and 30 tables regular transactional replication against a very frequently used database. My question is that 1. It's transactional replication hit a lot of performance? 2 Could we use snap-shot replication to reduce the performance issue. 3. Indexed-view really needs transactional replication? Thanks,
From: Hilary Cotter on 19 Sep 2006 10:44 1) that depends on your load and your hardware. However on some of the high throughput systems I have worked on the performance hit is seldom above 10-15%. Note that replicating text has a more severe performance impact. 2) Yes, but then your latency is much greater and you have to send all your data each time you run your snapshot - I think you will find that snapshot replication is not a scalable solution when you only need to replicate the changes. 3) No, you can replicated indexed views as schema only objects using snapshot replication. However if you want to replicate an indexed view to a table and replicate the changes you need snapshot replication. -- Hilary Cotter Director of Text Mining and Database Strategy RelevantNOISE.Com - Dedicated to mining blogs for business intelligence. This posting is my own and doesn't necessarily represent RelevantNoise's positions, strategies or opinions. Looking for a SQL Server replication book? http://www.nwsu.com/0974973602.html Looking for a FAQ on Indexing Services/SQL FTS http://www.indexserverfaq.com "mecn" <mecn2002(a)yahoo.com> wrote in message news:e1SI%23N$2GHA.4972(a)TK2MSFTNGP03.phx.gbl... > Hi all, > > We are planning to have transactional repllication for 8 > tables(indexed-view rep) and 30 tables regular transactional replication > against a very frequently used database. > My question is that 1. It's transactional replication hit a lot of > performance? 2 Could we use snap-shot replication to reduce the > performance issue. > 3. Indexed-view really needs transactional replication? > > Thanks, >
From: mecn on 19 Sep 2006 11:53
Hilary, It really helps us.Thanks a lot. I appreciate it. Henry "Hilary Cotter" <hilary.cotter(a)gmail.com> wrote in message news:es$oan$2GHA.3476(a)TK2MSFTNGP04.phx.gbl... > 1) that depends on your load and your hardware. However on some of the > high throughput systems I have worked on the performance hit is seldom > above 10-15%. Note that replicating text has a more severe performance > impact. > 2) Yes, but then your latency is much greater and you have to send all > your data each time you run your snapshot - I think you will find that > snapshot replication is not a scalable solution when you only need to > replicate the changes. > 3) No, you can replicated indexed views as schema only objects using > snapshot replication. However if you want to replicate an indexed view to > a table and replicate the changes you need snapshot replication. > > -- > Hilary Cotter > Director of Text Mining and Database Strategy > RelevantNOISE.Com - Dedicated to mining blogs for business intelligence. > > This posting is my own and doesn't necessarily represent RelevantNoise's > positions, strategies or opinions. > > Looking for a SQL Server replication book? > http://www.nwsu.com/0974973602.html > > Looking for a FAQ on Indexing Services/SQL FTS > http://www.indexserverfaq.com > > > > "mecn" <mecn2002(a)yahoo.com> wrote in message > news:e1SI%23N$2GHA.4972(a)TK2MSFTNGP03.phx.gbl... >> Hi all, >> >> We are planning to have transactional repllication for 8 >> tables(indexed-view rep) and 30 tables regular transactional replication >> against a very frequently used database. >> My question is that 1. It's transactional replication hit a lot of >> performance? 2 Could we use snap-shot replication to reduce the >> performance issue. >> 3. Indexed-view really needs transactional replication? >> >> Thanks, >> > > |