Adding more memory = hugh cpu load - Mailing list pgsql-performance

Hi,

Yesterday, a customer increased the server memory from 16GB to 48GB.

Today, the load of the server hit 40 ~ 50 points.
With 16 GB, the load not surpasses 5 ~ 8 points.

The only parameter that I changed is effective_cache_size (from 14 GB to
40GB) and shared_buffers (from 1 GB to 5 GB). Setting the values back
does not take any effect.

This server use CentOS 5.5 (2.6.18-194.3.1.el5.centos.plus - X86_64).
Should I change some vm parameters to this specific kernel ?

Thanks for any help.