What is the problem and a valid solution?

By:
In: SAP-C01
Tagged:
With: 2 Comments
You have launched an EC2 instance with four (4) 500 GB EBS Provisioned IOPS volumes attached.The EC2 instance is EBS-Optimized and supports 500 Mbps throughput between EC2 and EBS.The four EBS volumes are configured as a single RAID 0 device, and each Provisioned IOPS volume is provisioned with 4,000IOPS (4,000 16KB reads or writes), for a total of 16,000 random IOPS on the instance.The EC2 instance initially delivers the expected 16,000 IOPS random read and write performance.Sometime later, in order to increase the total random I/O performance of the instance, you add an additional two 500 GB EBS ProvisionedIOPS volumes to the RAID.Each volume is provisioned to 4,000 IOPs like the original four, for a total of 24,000 IOPS on the EC2 instance.Monitoring shows that the EC2 instance CPU utilization increased from 50% to 70%, but the total random IOPS measured at the instance level does not increase at all.

What is the problem and a valid solution?

The EBS-Optimized throughput limits the total IOPS that can be utilized; use an EBSOptimized instance that provides larger throughput.

Small block sizes cause performance degradation, limiting the I/O throughput; configure the instance device driver and filesystem to use 64KB blocks to increase throughput.

The standard EBS Instance root volume limits the total IOPS rate; change the instance root volume to also be a 500GB 4,000 Provisioned IOPS volume.

Larger storage volumes support higher Provisioned IOPS rates; increase the provisioned volume storage of each of the 6 EBS volumes to 1TB.

RAID 0 only scales linearly to about 4 devices; use RAID 0 with 4 EBS Provisioned IOPS volumes, but increase each Provisioned IOPS EBS volume to 6,000 IOPS.

Explanations:

The throughput limit of the EBS-Optimized instance can cap the IOPS that can be utilized, so switching to an instance type with higher EBS-Optimized throughput is a valid solution to utilize the additional IOPS from the new volumes.

While block size can affect I/O performance, simply changing to 64KB blocks does not address the underlying issue of hitting throughput limits. The problem isn’t solely related to block size configuration.

The root volume being standard EBS does not limit the IOPS rate. The IOPS limit is based on the total provisioned IOPS across all volumes, not just the root volume.

Increasing the volume size to 1TB does not inherently increase IOPS unless the volume is provisioned for a higher IOPS rate. Simply increasing size may not resolve the issue if the instance is still limited by throughput.

RAID 0 can support more than 4 devices, and the issue is not with the number of devices in the RAID setup but rather with the instance’s throughput limitations. Increasing IOPS without addressing throughput will not improve performance.

2026-04-07

2 Comments

  1. Stephanie
    Author

    If I had to guess, I’d say the answer is:
    The EBS-Optimized throughput limits the total IOPS that can be utilized; use an EBSOptimized instance that provides larger throughput.

  2. Diana
    Author

    It appears that the answer is:
    The EBS-Optimized throughput limits the total IOPS that can be utilized; use an EBSOptimized instance that provides larger throughput.

Leave a Reply

Your email address will not be published. Required fields are marked *

fourteen − nine =