I am considering upgrading my Postgres instances. I’m wondering if anyone has made any benchmarks to determine the benefits of using multiple shared CPUs for a Postgres cluster.
Are there any advantages to going above 2x shared CPUs?
What is the purpose? If you use Parallel Query, for example, such a choice might make sense. However, using PgBouncer or similar for connection pooling is a more practical solution. (All this is workload-specific, but often a good number of connections to have is roughly 3-5 times the CPU core count.) If you’re planning to use it in a production environment, I would suggest combining the horizontal scale of the shared machine with PgBouncer. However, sometimes shared machines can have a high steal time, so if you’re not sure you can control it nicely, the performance machine may be a good choice.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.