3.2 Examining CloudWatch metrics
Examining CloudWatch Metrics for Volume Gateway
- From the AWS console, click Services and select EC2.
- Select Instances from the menu on the left.
- Click on the Volume Gateway instance
- On the bottom panel click Monitoring and you should see the following:
- Click on CPU Utilisation and then View in metrics to display a more in-depth graph. This will open a new CloudWatch Management Console tab.
- Change the statistic to Maximum, time range to Last Hour and period to 1 minute. You can highlight the graph area to zoom in on the detail.
- We can see the first peak of around 40% - 50% CPU is where we completed the read test, the second peak between 90% - 100% is where we did the write test which indicates the write test is CPU constrained.
- While still in the CloudWatch Metric tab, on the left hand side select All metrics tab and then EC2 -> Per-Instance Metrics then choose your Volume Gateway from the list of EC2 instances.
- Select EBS writeops and EBS readops and set to graph to 30 min time range under Custom.
- Then choose Graphed metrics tab and set statistic to Maximum and period to 1 second.
- We can see in the read test a clear peak which has a straight line indicating that the read test is likely disk constrained from the cache disk.
Note: your results may vary. This exercise is to give an idea of how you can monitor your Volume Gateway performance.
Module 3 Summary
In this module you have simulated a read and write test using FIO to determine the maximum throughput available on your Volume Gateway. You have then examined CloudWatch metrics to understand what is constraining this throughput.
In the next module you will format and write data to the Volume Gateway volume, then snapshot this and covert it to an EBS volume. This illustrates how Volume Gateway can be used for migration, disaster recovery or copy data to a development machine in AWS.
Continue to Module 4.