> As far as I can tell, pgPool makes a connection to each backend for > every connection on the front-end. That is, if you have a 4-server > cluster and allow 100 connections, you will have: > > client-pgpool: 100 proc > pgpool-db1: 100 proc > pgpool-db2: 100 proc > pgpool-db3: 100 proc > pgpool-db4: 100 proc > > Is this correct?
Yes. > If I'm load-balancing between db1-4, it seems like I don't need to have > a equal number of connections for each server. As long as pgpool-II is not operated in replication mode, that will be true. However you always need 100 proc in the master node. So I'm not sure that buys anything. -- Tatsuo Ishii SRA OSS, Inc. Japan English: http://www.sraoss.co.jp/index_en.php Japanese: http://www.sraoss.co.jp _______________________________________________ Pgpool-general mailing list [email protected] http://pgfoundry.org/mailman/listinfo/pgpool-general
