Pay a visit to Pit Row at a race like the Daytona 500 and you’ll find teams crawling around the stock cars carrying bolt blasters, socket wrenches and super-sensitive listening devices. In the data center, the database administrator setting up his first Postgres database does much the same kind of work though the hotrod is the data environment, the tools are configuration settings and the crawling around is done through a keyboard.
The goal is the same—firing up for optimal performance. And just like a stock car, the database needs some tuning up before it can hum and roar at the starting line.
Here we’ll explore what Postgres users encounter as they deploy new databases. This is the third installment of our blog series that explores the common challenges new Postgres users encounter and some the strategies we employ to help. EnterpriseDB has created bundles of targeted services and training to help organizations as they progress.
(You can find an overview of the blog series in the first installment here.)
Bottom of the Stack
Postgres has a long history of evolutionary development, so upgrades are designed to integrate into the existing code without creating edge cases. This has implications for hardware because Postgres continues to support older hardware even as the database advances. But if your hardware is new, you’ll need to take a proverbial wrench to the out-of-box settings to optimize Postgres’ performance with these newer machines. This can also mean digging into the operating system configurations for the hardware as well to make sure these three wheels are rolling in unison. Read the rest of this entry »Pit Row: Getting Your Postgres Database to the Starting Line