Monitor Amazon Cloud Compute (EC2)

This measures the percentage of allocated EC2 compute units that are currently in use on the instance. Tracking this metric helps ensure that the instance is sized appropriately for your workload. If you see degraded application performance alongside continuously high levels of CPU usage (without any accompanying constraints on network, disk I/O, or memory), CPU may be a resource bottleneck. You can review application-level data using Application Performance Monitoring and/or potentially switch to using an instance type with more vCPU.

Very low usage of CPU utilization across your EC2 instances (for example, less than 5% for multiple weeks) can also help you determine if your instances are underutilized for your workload and potentially can be replaced with a smaller instance type to save cost.

The T instance family provides a baseline CPU performance with the ability to burst above the baseline. These instances are ideal for applications that are not generally CPU intensive but may benefit from higher CPU capacity for brief intervals. Credits are earned any time the instance is running below its baseline CPU performance level.

Amazon EC2 CPU credit metrics help keep track of your available balance and usage so that you are aware of possible charges as a result of extended bursting. Monitoring your instances’ credit usage can help you identify if you might need to switch to an instance type that is optimized for CPU-intensive workloads.

A burstable performance instance configured as unlimited can sustain high CPU utilization for any period of time whenever required. T4g, T3a, and T3 instances launch as unlimited by default. If the average CPU usage over a 24-hour period exceeds the baseline, you incur charges for surplus credits.

The CPU surplus credit balance metric tracks the accumulated balance. The CPU surplus credit charged measures extra credits that will result in additional charges.

Disk read and disk write bytes measure the number of bytes read from and written to the instance store volumes attached to the EC2 instance. Monitoring these metrics can help you understand application-level issues. Too much consistent reading of data from disk may indicate that your application could benefit from a caching layer. Lots of disk write for a long term could mean request queuing and potential application slowdowns if the disk speed is not fast enough to match the use case.

Disk read bytes per second and disk write bytes per second provide the average read/write data size for the specified period of time.

Disk read/write operations count gives you insight into the average number of read/write operations, which can help you determine if degraded performance is the result of consistently high I/O operations per second (IOPS) causing bottlenecks as disk requests become queued. If your instance volumes are HDD, you can consider a move to faster SSD disks.

Disk read/write operations count per second provides the average read/write IOPS for the specified period of time.

These metrics report network throughput, in bytes, of your EC2 instance. Any network drops or fluctuations can potentially be correlated to the application metrics to find the root cause of possible issues. Any throughput limit may indicate that the instance type may be mismatched with the application needs, especially for data intensive applications.

When looking at a set of EC2 instance data, if you see a considerable difference in network traffic load, you may require a load balancer to distribute the load more evenly across the related instances and be able to achieve better performance.

This check reports whether the instance has passed both the instance status check and the system status check. This check returns 0 (passed) if an instance passes the system status check or 1 (failed) if it fails.

The Elastic Amazon EC2 integration also collects important cloud metadata like instance name (cloud.instance.name=elastic-package-test-33138), instance id (cloud.instance.id=i-0de58890d94dda2e3), account name (cloud.account.name=elastic-beats), account id (cloud.account.id=428152502467), provider (cloud.provider=aws), region (cloud.region=us-east-1), availability zone (cloud.availability_zone=us-east-1c), and instance type (cloud.machine.type=t1.micro). This metadata helps in grouping and filtering collected data and helps drive various usage and performance trends across a set of related EC2 instances.

The Elastic EC2 integration comes with a default dashboard that shows key metrics like status check, CPU utilization, Disk I/O, and network throughput. It also provides filtering for region, availability zone, and machine type.

This check monitors the software and network configuration of the instance. Problems that can cause instance status checks to fail may include: incorrect networking or startup configuration, exhausted memory, corrupted file system, incompatible kernel, and so on. When an instance status check fails, you typically must address the problem yourself. You may need to reboot the instance or make instance configuration changes. To troubleshoot instances with failed status checks, refer to the Amazon EC2 documentation.

This check returns 0 (passed) if an instance passes the system status check or 1 (failed) if it fails.

This status check detects underlying problems with your instance that require AWS involvement to repair. Problems that can cause system status checks to fail include: loss of network connectivity, loss of system power, and hardware/software issues on the physical host. For instances backed by Amazon EBS, you can stop and start the instance yourself, which in most cases results in the instance being migrated to a new host. For instances backed by an instance store, you can terminate and replace the instance. Note that instance store volumes are ephemeral, and all data is lost when the instance is stopped.

This check returns 0 (passed) if an instance passes the system status check or 1 (failed) if it fails.