Can Amazon RDS Small do 200 temporary updates on a primary key per second?

I am trying to figure out how much load the SMALL RDS instance type can handle.

I know that this is very subjective, but given that it runs with the main key and the timestamp field is updated, the table has about 500,000 rows, will this be a problem? Suppose this is the only table in the database.

If this is a problem, why do I need a large copy?

What is a reasonable estimate of how many such updates a small / large instance can handle?

+4
source share
1 answer

Short answer: Yes, easy.

Long answer:

  • Mysql 5.5
  • Innodb table with rows ~ 200k
  • 128-bit primary primary key
  • Datetime column is not part of any key
  • No other indexes
  • Otherwise, idle RDS m1.small instance
  • 20x mysql parallel commands executing UPDATE tbl SET date='$now' WHERE pk='$key' for different random shuffled lists of existing primary keys

This links instance processors to 100% and maintains a constant update rate of just over 2000 updates per second. I / O was written in approximately 600 IOPS. Since it is associated with the processor, if you use prepared statements or perform batch inserts, I am sure that you can do much better before you finish working with the processor.

+3
source

All Articles