1

One of my client is having RDS MySQL DB for which write IOPS is high when compared to read IOPS. Tried increasing CPU and Memory to 8 and 61 GB respectively, but still it is the same. This is a small e-commerce application, which is not so popular. I checked DB connections, it always moves from 5 - 20. I am not a DB expert, but I checked Innodb file size and like wise parameters and everything looks the default value.

Can someone give me an idea on what could be the issue - whether its something with respect to infrastructure or maybe because of bad coding/sql queries.? In either case are there any inputs to reduce this write IOPS.

FYI, DB size is 50 GB total with provisioned IOPS up to 1000. Single largest table size is ~ 500 MB.

Any help would be really appreciated.

DB Connections - DB Connections Write IOPS - Write IOPS read IOPS - read IOPS

Improve this question
3
  • 1
    Write IOPS are typically caused by writes... Insert/Update/Delete queries. There is no intrinsic reason why they should be higher or lower than reads. It's very workload dependent. But they can also be caused by poorly-written queries that require implicit creation of temporary tables. There's no magic bullet. You need to understand what the application is actually doing. Commented Jun 18, 2018 at 13:40
  • @serverstackqns Have you figure out issue? I am struggling with similar problem. Commented Feb 19, 2019 at 18:48
  • Which MySQL version are you running? Commented Feb 27, 2019 at 20:18

1 Answer 1

Reset to default
1

From my (still limited) understanding the IOPS numbers reflect disk I/O and not operations in general.

With this is mind I think the following applies:

  • Every write operation to the DB (i.e. any INSERT, UPDATE etc. statement) updates the data in memory and also triggers a write operation to disk (to persist the data)
  • Read operations (i.e. SELECT statements) however can mostly be served from memory (especially with smaller datasets) and therefore do not cause any disk read operations.
Improve this answer

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.