- Newest
- Most votes
- Most comments
Hello.
I think the following blog post will be helpful.
The workload of your application does not place a heavy load on RDS, so gp3 should work fine.
https://directeam.io/blog/rds-storage-types-which-one-should-i-choose/
https://cloudfix.com/blog/the-shift-from-io1-io2-to-gp3-rds-db-instance-volumes/
Also, according to the AWS blog below, latency may vary depending on the volume type.
https://aws.amazon.com/jp/blogs/storage/how-to-choose-the-best-amazon-ebs-volume-type-for-your-self-managed-database-deployment/
Performance certainly plays a very important role in selecting the most appropriate volume type for your workload. Understanding the characteristics of your workload requirements is critical to ensure your applications function efficiently for your business. According to AWS documentation latency differences between io2 (sub-millisecond) versus gp3 (single digit millisecond) may make a difference in your selection. While you may find that your gp3 volumes can deliver similar IOPS performance to io2 Block Express (as seen in our example above), the difference is that io2 and io2 Block Express deliver consistent sub-millisecond response times which can be key for mission critical workloads.
Based on your description, switching from io1 with 1000 IOPS to gp3 would likely be beneficial for your workload.
Amazon EBS gp3 volumes provide a baseline performance of 3,000 IOPS and 125 MiB/s throughput regardless of volume size, which is higher than your current io1 configuration at 1000 IOPS. Since your workload typically uses around 400 IOPS and only occasionally reaches the 1000 IOPS threshold, gp3's 3,000 IOPS baseline would comfortably handle your performance needs, including those short bursts that are currently hitting your limits.
From a cost perspective, gp3 volumes offer significant savings compared to io1 volumes. You can save up to 87% by choosing gp3 over io1/io2 volumes for workloads requiring less than 16,000 IOPS and 500 MiBps throughput, which your workload falls well within.
While io1/io2 volumes have traditionally been recommended for production workloads requiring consistent performance, gp3 volumes now offer excellent performance characteristics that are suitable for many production workloads. They provide single-digit millisecond latency and 99.8% to 99.9% volume durability.
If your application ever grows to need more performance, gp3 volumes allow you to scale IOPS (up to 16,000 for RDS) and throughput independently of storage size, giving you flexibility for future growth.
The only scenario where you might want to stay with io1 or consider io2 Block Express would be if your application absolutely requires sub-millisecond latency or the highest possible durability (99.999%). However, based on your description, gp3 should meet your performance needs while providing significant cost savings.
Sources
Capture and diagnose I/O bottlenecks on Amazon RDS for SQL Server | AWS Database Blog
Amazon EBS General Purpose SSD volumes - Amazon EBS
Maximizing Storage Performance and Savings with Amazon EBS gp3 and Ollion | AWS Partner Network (APN) Blog
Optimize storage for SQL Server - AWS Prescriptive Guidance
Optimize Amazon RDS performance with io2 Block Express storage for production workloads | AWS Database Blog
While this answer look good, I do not trust generative AI. By experience it only answers everything in a way that will twist to agree with the way the question was asked to make the user happy about themselves. I would like an opinion from a real human with knowledge in the matter.
Relevant content
- asked 2 years ago
- asked 2 years ago
- AWS OFFICIALUpdated 2 years ago
Thank you. The goal is to get better performance, not worsen them, but at a reasonable cost. So I do believe gp3 is better suited to my workload. You first article even have a mention "If you don’t need over 64,000 IOPS, don’t use IO1." The latency aspect is interesting, but I doubt the difference between sub-milli to milli will have as much impact as the main iops/throughput values.