Ahh ok, yeah both emails are very useful. I also thought about replication but because in the article is covered as a separated topic i thought that it had nothing to do with scalability.
Regards and thanks.
*************************** Oscar Calderon Analista de Sistemas Soluciones Aplicativas S.A. de C.V. www.solucionesaplicativas.com Cel. (503) 7741 7850 Tel. (503) 2522-2834
http://reshmaparveen.blogspot.com/2009/12/postgresql-system-architecture.html
I was reading the Scalability section, where it mentions this:
Even though the query receiving thread is alone it still offers better or
equal scalability to MySQL. In terms of multi-computer scalability,
PostgreSQL does not scale at all.
This statement was wrong in 2009 and it's still wrong today. We were
using slony well before 2009 with read slaves to handle massive read
loads. While muti-master setups are still pretty new in the PostgreSQL
universe, there are some seups like Bucardo. Of course this paper
doesn't mention whether or not they're referring to shared storage or
separate storage, and what kind of loads would be expected. RedHat
Cluster server can provide failover etc. There are several different
options that pre-date this article.
The fact that it then goes on the sing the praises of MySQL clusters
as reliable and stable makes me question the whole article.
And, the page mentioned above seems to be copied verbatim (and without attribution) from a 2004 paper here:
