Robert Scott wrote: > > This is how the postgres 9 replication works. The replicating servers > become "hot standbys" which you can use for read requests. So in theory > the read requests could be scaled quite easily once set up. Atomicity of > the API would potentially suffer though. > One random thought of why it might be beneficial to not (exclusively) use the postgres replication is that you might not want to replicate all the db across all of your read only mirrors. The current full db is about 1.5Tb in size. If I remember correctly, loading the db from a planet dump without history was only about 300Gb (although it has probably grown by now). This possibly starts getting into the regime where it might not be entirely infeasible to use SSDs to speed up db access.
Kai -- View this message in context: http://gis.638310.n2.nabble.com/scaling-tp5905305p5909278.html Sent from the Developer Discussion mailing list archive at Nabble.com. _______________________________________________ dev mailing list [email protected] http://lists.openstreetmap.org/listinfo/dev
