A company runs a media transcoding application on a fleet of Amazon EC2 `c5.xlarge` instan… — AWS Certified Solutions Architect - Professional

A company runs a media transcoding application on a fleet of Amazon EC2 `c5.xlarge` instances. Each instance has an attached $300\text{ GiB}$ Amazon EBS `gp3` volume used for caching intermediate video files during processing. During peak hours, transcoding jobs take twice as long to complete. Amazon CloudWatch metrics indicate that the EC2 instances have low CPU utilization (under $20\%$ ) but high I/O wait times, and the EBS volumes are constantly operating at $125\text{ MB/s}$ throughput. A solutions architect needs to optimize the compute and storage tiers so that each instance can sustain a target throughput of $500\text{ MB/s}$ . Which of the following architectural changes will resolve the performance bottleneck and meet the throughput requirement?

Upgrade the EC2 instances to `c5.9xlarge` to increase the dedicated EBS bandwidth, and modify the `gp3` volumes to provision $500\text{ MB/s}$ of throughput.Answer
B
Deploy an Application Load Balancer (ALB) in front of the EC2 instances, request AWS Support to pre-warm the ALB to handle the traffic spikes, and modify the `gp3` volumes to provision $500\text{ MB/s}$ throughput.
C
Migrate the temporary file storage to an Amazon RDS PostgreSQL Multi-AZ DB instance, and configure the application to write files to the primary instance and read them from the standby instance to distribute the load.
D
Configure the EC2 Auto Scaling Group to launch additional `c5.xlarge` instances based on EBS disk queue depth, and set the scaling cooldown period to $15\text{ seconds}$ to ensure rapid scaling response.

Answer

Upgrade the EC2 instances to `c5.9xlarge` to increase the dedicated EBS bandwidth, and modify the `gp3` volumes to provision

500\text{ MB/s}

of throughput.

The correct answer correctly identifies that EBS performance is bound by both the storage volume configuration and the EC2 instance's dedicated EBS bandwidth. The `c5.xlarge` instance is limited to

1,250\text{ Mbps}

156\text{ MB/s}

) of EBS throughput, meaning that configuring the storage volume alone cannot achieve the target performance. Upgrading the instance to a `c5.9xlarge` (

7,000\text{ Mbps}

/ ~

875\text{ MB/s}

) provides the necessary host bandwidth, and modifying the `gp3` volume to

500\text{ MB/s}

removes the volume-level throughput limit.

Step-by-Step Solution

Identify the storage tier bottleneck.

The `gp3` volume has a default throughput baseline of

125\text{ MB/s}

. This matches the observed performance limit in CloudWatch.

Before upgrading compute, we must determine if the storage volume itself is the primary limitation.

Evaluate the compute tier EBS-optimized bandwidth limit.

A `c5.xlarge` instance has a dedicated EBS bandwidth limit of

1,250\text{ Mbps}

(approximately

156\text{ MB/s}

Even if the `gp3` volume is provisioned to

500\text{ MB/s}

, a `c5.xlarge` instance will throttle the storage traffic to

156\text{ MB/s}

due to instance-level limitations.

Select a compatible EC2 instance type and size.

A `c5.9xlarge` instance supports a dedicated EBS bandwidth of

7,000\text{ Mbps}

(approximately

875\text{ MB/s}

), which is greater than the

500\text{ MB/s}

(

4,000\text{ Mbps}

) requirement.

Upgrading the instance size is necessary to align the compute interface capability with the storage requirements.

Modify the EBS volume configuration.

Provision the `gp3` volume to deliver

500\text{ MB/s}

of throughput.

This raises the volume performance limit to the target rate, which is now supported by the upgraded instance size.

Key Concept

EC2 instance-level EBS bandwidth limitations and gp3 throughput configuration

Question

Answer

Step-by-Step Solution

Key Concept

Common Mistakes

Topics