AWS RDS CPU Utilisation 100%

Question

I have a java-based JPA2/hibernate web app deployed using Elastic Beanstalk that connects to an EC2 instance and RDS MySQL DB, used to be on micro and small classes. This was until the latency was unbearable, took like 30 seconds to load a page. Then i went to poke around the metrics and saw 100% cpu utilisation bottlenecks on the AWS RDS DB instance. Currently only having less than 10 users concurrently using it. So I went ahead to upgrade them to m3.large classes.

The first few minutes were fast as lightning but soon choked again. I went to take a look at the config monitoring for my elastic beanstalk and it was 3% CPU utilisation but my RDS CPU utilisation reaches 100% very quickly and once again takes roughly 30 seconds to load a single page.

This is my first experience using AWS's services so I'm really clueless as I heard it was supposed to be much better than this.

Does anyone know whats going on? Or how i can debug this? Am I making too many calls to the DB when using hibernate thus choking it?

Answer 1

It doesn't sound like an AWS RDS issue, more like something wrong on the MySql configuration / usage. How many records do you have there? Does it do a FT search?. Please try to be more specific.

Jonathan