Because Amazon Web Services (AWS)
offers so many different services and products, they can sometimes be difficult to understand and manage. This is especially true with the complex variety of analytics and monitoring reports provided by AWS that are available to users of their cloud computing platform. But it doesn’t have to be this way! There are plenty of ways to understand these reports and monitor your usage to ensure that you are getting the most out of your AWS resources, whether they are on-demand instances or reserved with Spot Instances.Introducing Amazon CloudWatch
Amazon CloudWatch is a monitoring service for AWS, which provides metrics on your AWS resources at regular intervals.
For example, you can keep track of how much disk space your servers are using. It also has alarms that send you notifications if certain thresholds are met.
This makes it easy to stay on top of any potential problems before they get out of hand. And Amazon Web Services (AWS) already comes with more than 15 built-in alarms, giving you a great place to start building an automated management system for your business.
However, if none of these automatically suits your needs, it’s simple enough to create custom alarms as well! To do this, just pick the metric you want to monitor and set the threshold level you want the alarm to trigger when that metric reaches or exceeds the limit.
You’ll receive a notification via email or SMS text message when the threshold is reached. If desired, you can also have the alarm trigger other actions like sending notifications or making calls.
Introducing Amazon CloudTrail
Amazon CloudTrail is a web service that AWS customers can use to track their servers’ activity. You can identify which users have performed activities on specific resources in your account (such as creating, reading, updating, or deleting an object) by tracking events related to your servers such as API calls, changes to permissions, and the creation of new instances.
CloudTrail generates log files that you can view in Amazon S3 or download into an Amazon S3 bucket of your choice.
From there you can load them into any third-party system or analyze them with other AWS services like Lambda@Edge or Athena.
Some people may be using Amazon Kinesis Data Streams or Amazon EMR for this purpose. Either way, this would involve the ability to easily parse JSON logs, which is something ELK does very well.
It is also possible to set up alerts based on triggers generated from CloudTrail logs—for example, if someone creates a new user in IAM without proper authorization and you are set up for that event type.
Introducing the Amazon Trusted Advisor Tool
If you’re having a tough time interpreting all of these reports, you may be interested in Amazon Trusted Advisor.
This tool analyzes your account and shows you how you can cut costs and improve performance—for free.
Amazon Trusted Advisor even gives specific recommendations on ways to do so. Think of it as a personal assistant that keeps tabs on your AWS usage, then figures out how to make things more efficient for you.
It currently only runs in Amazon EC2, but we expect it’ll spread throughout other services soon enough. The site is also a great way to learn about the new features that are released every day on AWS.
Some features include Elastic File System (EFS), Serverless Application Model (SAM), Elastic Load Balancing, and more.
Introducing Amazon GuardDuty
Amazon GuardDuty can provide deep visibility into unusual activities in your AWS environment. It’s designed for IT teams that operate at scale, which is why it can automatically collect data from your AWS resources.
This means you don’t have to spend time writing or configuring rules or alerts. Amazon GuardDuty also uses advanced machine learning algorithms and threat intelligence feeds, so you can quickly see what events are important to you, even when there are thousands of them every day.
Monitoring Security – You can use the visualization engine to display security incidents on a map.
Alert Management – You can set up notifications for all your monitored resources (called tags) using Amazon GuardDuty’s simple rule language.
Compliance Overview – You can generate reports based on compliance requirements and export these reports in either CSV or PDF format.
Introducing Amazon Inspector
Amazon Inspector is a new service that helps you continuously monitor your AWS accounts for configuration vulnerabilities.
These issues might represent breaches in security or opportunities for you to save money, both of which are important.
Amazon Inspector works as an agent on your EC2 instances and then provides actionable information about any potential issues it finds.
It will help you find out how protected your account is so that you can fix issues before they become problems (in either regard).
Amazon Inspector uses automated security assessment techniques (i.e., dynamic analysis) that identify weaknesses in configurations by executing attack code, just like real attackers do – only with permission from Amazon engineers who have reviewed and approved these tools before use.
You don’t need to download, install, or maintain any software on the servers where Amazon Inspector runs; all the work is done remotely by the Inspector agent.
You can even take snapshots of what’s happening in your environment during a scan if you want to see exactly what has been done.
Why You Need to Track These Services
Amazon Web Services (AWS) offers many services, each with its dashboard. You can use these dashboards to monitor your services’ performance.
On a given day, you may be using multiple AWS services (CloudFront for CDN, S3 for object storage, Route 53 for DNS management).
Since each service reports on different metrics and most of them don’t necessarily correlate, you must be able to look at all of them in one place.
That way you can see how your various AWS resources are performing relative to one another. This is particularly useful if you want to understand why some of your CloudFront requests seem slow today even though S3 is meeting its targets.
Understanding the Difference in Metrics
As we mentioned earlier, a metric is simply a single data point. For example, you can use Amazon CloudWatch to monitor average CPU utilization or uptime percentage.
A metric is often used interchangeably with a statistic (which is just a value derived from many samples of data).
When it comes to reporting, you’ll sometimes see one without the other. Many times, reports will simply state operational metrics or statistics when referring to both together.