IOPS is a difficult metric to 'improve'; unless you are in control of the code that's hammering your disk and can optimize its disk access patterns, there's not much you can do with this. Many IOPS typically equates to heavy system usage for a database or web server - which isn't necessarily a bad problem to have. Is there a reason you want to bring this number down?
We had a very large database in their biggest RDS instance and just got hammered with large IO spikes.. We just moved off RDS to a EC2 instance and things got better.
IOPS is a difficult metric to 'improve'; unless you are in control of the code that's hammering your disk and can optimize its disk access patterns, there's not much you can do with this. Many IOPS typically equates to heavy system usage for a database or web server - which isn't necessarily a bad problem to have. Is there a reason you want to bring this number down?
We had a very large database in their biggest RDS instance and just got hammered with large IO spikes.. We just moved off RDS to a EC2 instance and things got better.
Could it be the slow query log or general log doing lots of writes? See here.