I've not done a formal study, but I've always found that throwing hardware at the problem does wonders. My current database I made faster by bumping RAM until the entire working set fits in memory. The server has 256GB of RAM, half of which is used by ZFS for its purposes, and the other half for Postgres. The prior iteration of the servers only had 64GB of RAM and the difference was very remarkable.
On Sat, Apr 22, 2017 at 11:27 AM, Melvin Davidson <melvin6925@xxxxxxxxx> wrote:
Has anyone ever done a study on performance increase via ram increase?
I have a client on AWS with 8GB total ram (2GB shared_buffers), and I am
curious if doubling the ram to 16GB (4GB shared_buffers) will result in
minimizing query response time.
--Melvin Davidson
I reserve the right to fantasize. Whether or not you
wish to share my fantasy is entirely up to you.
