---
title: 'Monitor executions with Amazon CloudWatch Metrics'
weight: 143

---

Monitoring metrics is important to maintain the reliability, availability, and performance of your workflows. 

This workshop deploys a state machine called *DetectSentimentStateMachine* and few other resources.

   ![DetectSentiment State Machine](/static/img/module-12/state-machine.png)

This state machine accepts an input string, detects the sentiment of the text in the string, and records the result of the analysis in an Amazon DynamoDB table. The workflow invokes a Lambda function which calls Amazon Comprehend to perform the sentiment analysis. This state machine is triggered once per minute by an Amazon EventBridge rule.

In this exercise, you will use CloudWatch Metrics to monitor the *DetectSentimentStateMachine* workflow executions.

The following Step Functions Execution metrics are available in CloudWatch 
- ExecutionTime	
- ExecutionThrottled
- ExecutionsAborted	
- ExecutionsFailed	
- ExecutionsStarted	
- ExecutionsSucceeded	
- ExecutionsTimedOut

More details about these metrics can be found [here](https://docs.aws.amazon.com/step-functions/latest/dg/procedure-cw-metrics.html#cloudwatch-step-functions-execution-metrics).

1. Navigate to [CloudWatch console](https://console.aws.amazon.com/cloudwatch/home) in your AWS console. Make sure you are in the correct region.

2. Under `Metrics` on the left navigation menu, click **All Metrics**. Find the metrics box labed **States** and click it.

   ![CW All Metrics States](/static/img/module-12/cw-all-metrics-states.png)

3. Click **Execution Metrics**.

   ![Execution Metrics](/static/img/module-12/cw-states-execution-metrics.png)

4. Select all the metrics listed for `DetectSentimentStateMachine`. 

   ![DetectSentiment Metrics](/static/img/module-12/cw-detect-sentiment-metrics.png)

5. Click the **Graphed metrics** tab. Update the `Statistic` column of `ExecutionTime` to **Average** and the `Statistic` for rest of the metrics to **Sum**. On the top right change the legend from Line to `Number`.

   ![Sum and Average](/static/img/module-12/cw-metrics-sum-avg.png)

6. Click the edit pencil next to the graph title,  type **Execution Metrics** and click **Apply**.

7. In the top right, click the **Actions** dropdown and choose **Add to dashboard**.

-  In the Add to dashboard page, click **Create new**.
- Enter *DetectSentiment* for the dashboard name and click **Create**.
- For **widget type**, select Number.
- Click **Add to dashboard**.

   ![CW Metrics](/static/img/module-12/cw-add-dashboard.png)

8. Choose **Save dashboard**.

You can now see the execution metrics for DetectSentiment State Machine Step Functions. You will notice there are executions that have failed, indicated by ExecutionsFailed metrics.

   ![Dashboard Metrics](/static/img/module-12/cw-dashboard.png)

In the next modules, you will use CloudWatch Logs and X-Ray tracing to debug and identify the root cause of these failures.