Which solution will meet these requirements MOST cost-effectively?
Double the gp2 EBS volume size from 500 GB to 1,000 GB.
Change the volume type from gp2 to General Purpose SSD (gp3).
Change the volume type from gp2 to Throughput Optimized HDD (st1).
Change the volume type from gp2 to Provisioned IOPS SSD (io2).
Explanations:
Doubling the volume size from 500 GB to 1,000 GB will increase the burst credits for IOPS, but it will not sustain 3,000 IOPS consistently. The gp2 volume type provides a maximum of 3,000 IOPS only if the size is 1 TB or more, but this is not the most cost-effective solution.
Changing from gp2 to gp3 allows you to independently provision IOPS and throughput. With gp3, you can provision up to 16,000 IOPS and 1,000 MB/s throughput, which will meet the 3,000 IOPS requirement without needing to increase volume size, making it a cost-effective solution.
Changing to st1 is not suitable for this use case. st1 is designed for throughput-intensive workloads like big data and log processing, not for workloads that require high IOPS like Jenkins builds. It provides lower IOPS compared to gp3 and doesn’t meet the 3,000 IOPS requirement.
Changing to io2 provides high IOPS, but it is a more expensive option meant for I/O-intensive workloads that require high durability and consistent performance. For the given requirement, gp3 is the most cost-effective solution, and io2 would result in unnecessary cost overhead.
I design that the answer is:
Change the volume type from gp2 to General Purpose SSD (gp3).
From my perspective, the answer is:
Change the volume type from gp2 to General Purpose SSD (gp3).