Amazon IoT Device Defender pricing

Detect allows you to continuously monitor high-value security metric data reported by your devices (e.g., the number of listening TCP/IP ports on your devices or list of IPs the device is communicating with) and the cloud (e.g., authorization failure count). When you use Rules Detect, the metric data is continuously evaluated against user-defined behaviors; when you use ML Detect, the metric data is continuously evaluated by automatically built machine-learning models to identify anomalies. In the case that an anomaly is detected, an alarm notification is sent to Amazon SNS.

When you turn on Audit, you are charged based on the number of active device principals in a month.
For certificates, we consider them as active when they were created, updated, or used to connect to Amazon IoT Core.
For principals other than certificates, we consider them as active when they were used to connect to Amazon IoT Core.

Rule-based Detect lets you define rules and thresholds to monitor devices in your fleet for operational and security issues across metrics like messages sent and connection attempts. You can also take built-in actions to respond to detected issues like quarantining a device. You are charged based on the number of metric datapoints reported to Amazon IoT Device Defender for monitoring. A metric datapoint, for example, is the number of messages sent by your device.

With Amazon IoT Device Defender, you will be billed separately for usage of Connectivity to Amazon IoT Core but will not incur additional Amazon IoT Core Messaging costs for metric datapoints reported to Amazon IoT Device Defender. You will be billed separately for alert notification delivery via Amazon SNS.

ML Detect provides an easy way to monitor devices in your fleet for operational and security issues across metrics like messages sent and connection attempts. You can also take built-in actions to respond to detected issues like quarantining a device. You are charged based on the number of metric datapoints reported to Amazon IoT Device Defender for monitoring. A metric datapoint, for example, is the number of messages sent by your device.

With ML Detect, metric datapoints are used for training ML models and performing anomaly evaluations. A charge will occur when metric datapoints are reported regardless whether ML Detect is ready to carry out anomaly evaluations or not.

You have 100 devices that connect to Amazon IoT Core every month. Your cost for Audit would be calculated as follows:

Charges = 100 devices X ¥0.0104 per device per month = ¥1.04 per month

You have 100 devices and they are reporting 4 metrics at the rate of 2 datapoints per hour with rule-based Detect turned on. Your cost for rule-based Detect would be calculated as:

Number of metric datapoints per month = 100 devices x 4 metrics x 2 datapoints per hour x 24 hours per day x 30 days per month = 0.576M

Charges = 0.576M metric datapoints x ¥0.24 per 100K metric datapoints = ¥1.38 per month

You have 20 devices and they are reporting 6 metrics at the rate of 2 datapoints per hour with ML Detect turned on. Your cost for ML Detect would be calculated as:

Number of metric datapoints per month = 20 devices x 6 metrics x 2 datapoints per hour x 24 hours per day x 30 days per month = 172.8K

Charges = 172.8K metric datapoints x ¥19.31 per 100K metric datapoints = ¥33.37 per month

If you increase your devices to 100 and they are reporting 6 metrics at the rate of 2 datapoints per hour with ML Detect turned on, your cost for ML Detect would be calculated as:

Number of metric datapoints per month = 100 devices x 6 metrics x 2 datapoints per hour x 24 hours per day x 30 days per month = 864K

Charges = 300K metric datapoints x ¥19.31 per 100K metric datapoints + (864-300)K metric datapoints x ¥7.24 per 100K metric datapoints = ¥98.76 per month

Your total monthly cost for Amazon IoT Device Defender is as follows:

Total monthly charges = ¥1.04 Audit cost + ¥1.38 Rule Detect cost + ¥98.76 ML Detect cost (for 100 devices reporting 6 metrics) = ¥101.18