AWS IoT SiteWise pricing
AWS IoT SiteWise is a managed service that makes it easy to collect, organize, process, and monitor data from industrial equipment at scale to help you make better, data-driven decisions. You pay only for what you use, with no minimum fees or mandatory service usage. You are charged separately for usage of Messaging, Data Processing, Data Storage, Data Export, AWS IoT SiteWise Monitor, AWS IoT SiteWise Edge, and Alarms.
AWS IoT SiteWise Pricing Tool
Estimate detailed cost for your AWS IoT SiteWise workload, including granular egress, ingestion, and computation pricing
Pricing details
AWS IoT SiteWise Messaging
Messages transport equipment data and computed data to, from, and within AWS IoT SiteWise. Messages are metered by the amount of data ingested or retrieved by querying or publishing.
Data Ingestion: Messages are ingested through an application programming interface (API)—a method allowing two software applications to exchange information, or generated by a computation over equipment data. A message can also be ingested through AWS IoT SiteWise APIs, including APIs used by AWS IoT SiteWise Edge or by other application clients (e.g., BatchPutAssetPropertyValue, CreateBulkImportJob).
For near real-time data ingestion, messages are metered in 1KB increments or 10 data points per data stream.
For buffered data ingestion, messages are metered in 5KB increments or 60 data points across multiple data streams. When ingesting buffered data, you will be charged a minimum of 10MB or 2,000 messages per ingestion period.
AWS IoT SiteWise integrates with Amazon Lookout for Equipment, to train anomaly detection models in Lookout for Equipment and predict failures before they happen. You will incur charges for publishing and receiving messages, computation, and storage in AWS IoT SiteWise.
Data Retrieval: Messages are retrieved when a property, such as a measurement of a sensor value, is published to a query request (e.g. when data is retrieved through APIs or published to AWS IoT Core).
Messages published in response to an asset property query request are metered for each set of up to 50 property values or every 4 KB of data contained in the message, whichever results in a higher count of messages. The warm tier retrieval meters a minimum of 200 KB or 10 messages per retrieval per data stream, measured in 20KB increments or 250 property values.
Messages published in response to an interpolated asset property query request are metered based on the number of property values retrieved from storage to compute each interpolated value.
Additional charges apply for GET requests made against your Amazon Simple Storage Service (Amazon S3) bucket when data results are retrieved from your cold tier. For more information about these charges, see Amazon S3 pricing. Messages published to AWS IoT Core are metered in 1 KB increments, and this operation does not incur any additional messaging charges from AWS IoT Core.
AWS IoT SiteWise Data Storage
Once data is in the cloud, AWS IoT SiteWise stores equipment and computed data in a managed time series database that automatically scales with your data usage and storage. AWS IoT SiteWise monitors the size of your storage continuously to determine your storage charges.
AWS IoT SiteWise Data Processing
AWS IoT SiteWise Query
Queries allow you to retrieve definition of asset models, asset hierarchies, assets, as well as measurements data.
You pay for the volume of data scanned. Data scanned, is metered across data stream to retrieve your results. Data scan meters metadata or telemetry data at increments of 100 KB per retrieval.
Additional charges for data egress from the hot and warm storage tiers apply. See Messaging pricing.
AWS IoT SiteWise Monitor
AWS IoT SiteWise Monitor provides managed web applications for interacting with operational data from devices and equipment that send data to AWS IoT SiteWise.
AWS IoT SiteWise Data Export
You can store data from measurements, metrics, transforms and aggregates in AWS IoT SiteWise cold storage tier. AWS IoT SiteWise stores data in cold storage tier that is older than your predefined hot storage tier retention period, including data that is ingested using BatchPutAssetPropertyValue and CreateBulkImportJob. Data in cold tier is stored as Amazon S3 objects in an S3 bucket in your account. You will be charged for the number of S3 objects created or updated by AWS IoT SiteWise to store data in cold tier.
Additional charges apply for storing exported data in Amazon S3 and for PUT requests made against your Amazon S3 bucket. For more information about these charges, see Amazon S3 pricing.
AWS IoT SiteWise Edge
AWS IoT SiteWise Edge provides on-premises data collection and processing capabilities. SiteWise Edge capabilities are offered through packs.
Data Collection Pack: This pack enables data collection from common industrial data sources and transferring it to the AWS Cloud. You can collect data from OPC-UA servers, Modbus servers, and over Ethernet/IP. You can transfer the collected data to Amazon S3, AWS IoT SiteWise, AWS IoT Core, Amazon Timestream, and Amazon Kinesis.
The Data Collection Pack is free.
Data Processing Pack: This pack enables the processing of the computations (transforms and metrics) with AWS IoT SiteWise asset models and visualizing them with SiteWise Monitor web applications on the SiteWise gateway. You can also interact with the data processed locally by the SiteWise gateway using the same REST APIs you use with AWS IoT SiteWise in the cloud.
A Data Processing Pack is charged at $200 per active gateway per month. A gateway is considered active in a month if it connects to the AWS Cloud to receive SiteWise Edge software configuration updates.
Note that you will be charged separately for any other AWS services that you use; refer to specific pricing for those services. You will also be charged separately for AWS IoT Greengrass, which is used by SiteWise Edge.
AWS IoT SiteWise Alarms
The Alarms feature allows you to set up, visualize, and manage rule-based alerts to assess equipment behavior and identify equipment performance issues. Alarms in AWS IoT SiteWise are supported through the Alarms feature of AWS IoT Events. You will incur charges for alarms you create in AWS IoT SiteWise through AWS IoT Events. See AWS IoT Events pricing for details.
You will incur charges in AWS IoT SiteWise for publishing messages to AWS IoT Events that are used to evaluate alarms. You will also incur charges in AWS IoT SiteWise for storing alarm states received from AWS IoT Events.
Pricing examples
-
Example 1
An industrial production line is sending data to AWS IoT SiteWise in the US East (N. Virginia) region. The production line includes 10 pieces of equipment, each of which reports one measurement every second. You configure AWS IoT SiteWise to collect this measurement data and upload to AWS. The monthly AWS IoT SiteWise charges are estimated as follows:Usage SummaryMessagingNumber of measurements per equipment = one measurement per equipment
Frequency of message ingestion = one message per second per measurement
Number of messages ingested in a month = one message / second / measurement * one measurement / equipment * 10 equipment * 86,400 seconds / day * 30 days / month = 25,920,000 messages per month
Total number of messages ingested per month = 25,920,000 messages per month = 25.92 million messages per month
Data Processing
Auto-computed aggregates
Number of aggregate computations per message = one computation per message ingested
Total number of messages ingested per month = 25,920,000 messages per month
Total number of aggregate computations per month = one computation / message * 25,920,000 messages / month = 25,920,000 computations = 25.92 million computations per monthTotal computations per month = 25.92 million computations per month
Data Storage
Equipment data
Number of measurement messages per month = 25,920,000 messages per month
Size of each message = 150 bytes per message (assuming each message ingested is 150 bytes)
Storage per month = 25,920,000 messages per month * 150 bytes / message * 1 GB / (1024^3 bytes) = 3.62 GB per month*Auto-computed aggregates
SiteWise stores all auto-computed aggregates. The storage used by auto-computed aggregates depends on the number of aggregates and time intervals over which they are calculated. In the example above, six auto-computed aggregates calculated over three time intervals (one-minute, one-hour, one-day) will result in 0.37 GB per month of data stored.Storage from auto-computed aggregates = 0.37 GB per month*
Total data stored per month = 3.62 GB (from Equipment data) + 0.37 GB (from Auto-computed aggregates) = 3.99 GB per month*
* This data reflects the data storage volume at the end of the first month.
Billing SummaryMessagingMessaging charges = $1.00 per million messages
Total messaging usage in a month = 25.92 million messages
Monthly charges for messaging = 25.92 million * $1.00 per million messages = $25.92 per monthData Processing
Data processing charges = $0.50 per million computations
Total computations in a month = 25.92 million computations
Monthly charges for computations = 25.92 million computations * $0.50 per million computations = $12.96 per monthData Storage
Total data stored at the end of the month = 3.99 GB per month
Average storage usage per hour during the month = 1.995 GB*
Monthly charges for storage = 1.995 GB * $0.30 per GB-month = $0.60 per month**
*Assuming storage has grown linearly from 0GB to 3.99GB by the end of the month, average storage usage would approximate 1.995 GB.
**Data that remains in storage from prior billing cycles will continue to be metered and billed accordingly
Total Monthly Bill
Monthly AWS IoT SiteWise charges = $25.92 + $12.96 + $0.60 = $39.48 per month
-
Example 2
On top of example 1, say you now also define 1 custom metric per equipment aggregated over 1-minute interval to monitor the performance. The additional monthly usage and charges are estimated as follows:Usage Summary
MessagingNumber of metrics per equipment = one metric per equipment
Frequency of messages generated per metric = one message / minute / metric
Number of messages generated in a month = one message / minute / metric * one metric /equipment * 10 equipment * 1440 minutes / day * 30 days / month = 432,000 messages per month
Additional number of messages ingested per month = 432,000 messages per month = 0.432 million messages per month
Data Processing
Custom metrics
Frequency of computation per metric = one computation / message * one message / second / equipment * one equipment / metric = one computation / second / metric
Number of computations per metric per month = one computation / second / metric * 86,400 seconds / day * 30 days / month = 2,592,000 computations per metric per month
Number of computations across all equipment per month = 2,592,000 computations / metric / month* one metric / equipment * 10 equipment = 25,920,000 computations = 25.92 million computations per month
Auto-computed aggregates
Additional number of aggregate computations per month = one computation / message * 432,000 messages / month = 432,000 computations = 0.432 million computations per month
Additional number of computations per month = 25.92 million (from Custom metrics) + 0.432 million (from Auto-computed aggregates) = 26.352 million computations per month
Data Storage
Custom metrics
Additional messages ingested per month = 432,000 messages per month
Size of each message = 150 bytes
Additional storage per month = 432,000 messages / month * 150 bytes / message * 1 GB / (1024^3 bytes) = 0.06 GB per month
Auto-computed aggregates
Additional storage from auto-computed aggregates on custom metrics = 0.37 GB.
Additional data stored per month = 0.06 GB (for Custom metrics) + 0.37 GB (from Auto-computed aggregates) = 0.43 GB per month
Billing Summary
Messaging
Additional messaging usage per month = 0.432 million messages
Additional monthly charges for messaging = 0.432 million messages * $1.00 per million messages = $0.43 per month
Data Processing
Additional computations per month = 26.352 million computations
Additional monthly charges for computations = 26.352 million computations * $0.50 per million computations = $13.18 per month
Data Storage
Additional data stored at the end of the month = 0.43 GB
Average additional storage usage per hour during the month = 0.25 GB*
Additional monthly charges for storage = 0.25 GB * $0.30 per GB-month = $0.08 per month**
*Assuming storage has grown linearly from 0GB to 0.43GB by the end of the month, average storage usage would approximate 0.25GB.
**Data that remains in storage from prior billing cycles will continue to be metered and billed accordingly
Additional Monthly Bill for custom metrics
Additional monthly bill = $0.43 + $13.18 + $0.08 = $13.69 per month
-
Example 3
On top of example 2, say you now also define one custom transform per measurement to do unit conversion (e.g., convert temperature from Celsius to Fahrenheit). The additional monthly usage and charges are estimated as follows:Usage Summary
MessagingNumber of transforms per measurement = one transform per measurement
Frequency of message ingestion per transform = one message / second / measurement * one measurement / one transform = one message / second / transform
Number of messages generated per month = 10 equipment * one measurement / equipment * one transform / measurement * one message / second / transform * 86400 seconds / day * 30 days / month = 25,920,000 messages per month
Additional number of messages ingested per month = 25,920,000 messages per month = 25.92 million messages per month
Data Processing
Custom transforms
Number of computations per month per transform = one computation / message * one message / second / measurement * one measurement / transform * 86,400 seconds / day * 30 days / month = 2,592,000 computations per month per transform
Number of computations across all equipment per month = 2,592,000 computations / month / transform * one transform / measurement * one measurement / equipment * 10 equipment = 25,920,000 computations per month = 25.92 million computations per month
Auto-computed aggregates
Additional number of aggregate computations per month = one computation / message * 25,920,000 additional messages / month = 25,920,000 computations = 25.92 million computations per month
Additional number of computations per month = 25.92 million (from Custom transforms) + 25.92 million (from Auto-computed aggregates) = 51.84 million computations per month
Data Storage
Custom transforms
Additional messages ingested per month = 25,920,000 messages
Size of each message = 150 bytes
Additional storage per month = 25,920,000 * 150 bytes / (1024^3 bytes/GB) = 3.62 GB per month
Auto-computed aggregates
Additional storage from auto-computed aggregates on custom transforms = 0.37 GB.
Additional data stored per month = 3.62 GB (for Custom transforms) + 0.37 GB (from Auto-computed aggregates) = 3.99 GB per month
Billing Summary
Messaging
Additional messaging usage per month = 25.92 million messages
Monthly charges for messaging = 25.92 million messages * $1.00 per million messages = $25.92 per month
Data Processing
Additional computations per month = 51.84 million computations
Monthly charges for computations = 51.84 million computations * $0.50 per million computations = $25.92 per month
Data Storage
Additional data stored at the end of the month = 3.99 GB
Average additional storage usage per hour during the month = 1.965 GB*
Additional Monthly charges for storage = 1.995 GB * $0.30 per GB-month = $0.60 per month**
*Assuming storage has grown linearly from 0GB to 3.99GB by the end of the month, average storage usage would approximate 1.995.
**Data that remains in storage from prior billing cycles will continue to be metered and billed accordingly
Additional Monthly Bill for custom transforms
Additional monthly charges = $25.92 + $25.92 + $0.60 = $52.44 per month
-
Example 4
To example 3, you also add graphs. Using your custom application, you create a dashboard consisting of two graphs to monitor your equipment performance. On one of these graphs, you are monitoring raw data for your measurement retrieved using the GETAssetPropertyValueHistory API. On the other graph, you are monitoring five-minute linearly interpolated data for another measurement retrieved using the GETInterpolatedAssetPropertyValues API. Your application refreshes these graphs every 30 seconds, and your graphs show data over a one-hour window. The additional monthly usage and charges are estimated as follows:Usage Summary
Messaging
Your SiteWise monitor dashboard automatically refreshes every five seconds to help ensure that you are visualizing data close to real time. Assuming that you are monitoring graphs over a one-hour window, whenever the dashboard refreshes, all data points plotted on that one-hour window get refreshed. So, for each graph showing a one-minute aggregate, 60 total data points get fetched upon every refresh through messages. As each of these messages can carry 50 data points, the total messages retrieved can be estimated as follows:
Number of messages retrieved per refresh per graph = ceiling ((60 data points / refresh / graph) / (50 data points / message)) = two messages per refresh per graph
Number of messages retrieved across all graphs per refresh = two messages / refresh / graph * 10 graphs = 20 messages per refresh
Number of messages retrieved across all users per month = (20 messages / refresh) * (one refresh / five seconds) * 3,600 seconds / hour * eight-hour shift / day / user * 30 days / month * five users = 17,280,000 messages per month
Additional number of messages retrieved per month = 17,280,000 messages per month = 17.28 million messages per month
SiteWise Monitor
Number of unique active users per month = five plant operators = five users per monthBilling Summary
Messaging
Additional messaging usage from SiteWise monitor per month = 17.28 million messages
Monthly charges for messaging = 17.28 million messages * $1.00 per million messages = $17.28 per monthSiteWise Monitor
SiteWise Monitor charges = $10.00 per active user per month
Number of active users per month = five users
Monthly charges for SiteWise Monitor = five users * $10.00 per active user per month = $50 per monthAdditional Monthly Bill for SiteWise Monitor
Additional monthly charges = $17.28 + $50.00 = $67.28 per month
-
Example 5
To extend example 4, you also add graphs using your custom application. Using your application, you create a dashboard consisting of two graphs to monitor your equipment performance. On one of these graphs you are monitoring raw data for your measurement retrieved using the GETAssetPropertyValueHistory API. On the other graph, you are monitoring five-minute linearly interpolated data for another measurement retrieved using the GETInterpolatedAssetPropertyValues API. Your application refreshes these graphs every 30 seconds and your graphs show data over a one-hour window. The additional monthly usage and charges are estimated as follows:Usage SummaryMessaging
For the one-hour graph showing the raw measurement data, with measurements coming in at one data point per second, 3,600 total data points get fetched from storage upon every refresh. As each of these messages can carry 50 data points, the total messages retrieved can be estimated as follows:
Number of messages retrieved per refresh = ceiling ((3,600 data points / refresh / graph) / (50 data points / message)) = 72 messages per refreshNumber of messages containing raw data retrieved per month = (72 messages / refresh) * (one refresh / 30 seconds) * 3,600 seconds / hour * 24 hours / day * 30 days / month = 6,220,800 messages per monthFor the one-hour graph showing the five-minute linearly interpolated measurement data, 12 interpolated values are returned, which means 24 messages (two messages per linearly interpolated value) get fetched from storage per refresh.Number of messages containing interpolated data retrieved per month = (24 messages / refresh) * (one refresh / 30 seconds) * 3,600 seconds / hour * 24 hours / day * 30 days / month = 2,073,600 messages per monthTotal additional number of messages retrieved per month = 6,220,800 (for graph showing raw data) + 2,073,600 (for graph showing interpolated data) = 8,294,400 messages per month = 8.29 million messages per month
Billing SummaryMessaging
Additional messaging usage per month = 8.29 million messages
Monthly charges for messaging = 8.29 million messages * $1.00 per million messages = $8.29 per monthAdditional Monthly Bill for graphs in your custom application
Additional monthly charges = $8.29 per month
-
Example 6
You configure a single SiteWise Edge gateway to monitor performance of a workcell with minimum dependency on the network. The workcell comprises 10 pieces of equipment. You model equipment in AWS IoT SiteWise with 10 transforms and two metrics (of one-minute intervals). You configure AWS IoT SiteWise to compute the transforms and metrics in the gateway, and forward the final result to the AWS IoT SiteWise cloud service for storage. You create a SiteWise Monitor dashboard to view the metrics for the workcell and configure it to run on the gateway.Usage SummarySiteWise EdgeNumber of packs Used:
One Data Collection Pack
One Data Processing Pack per month
MessagingFrequency of messages from transforms = one message per second per transform
Number of messages from transforms per month, per gateway = 10 transforms x one message per second x 86,400 seconds/day x 30 days = 25.92 million messages
Number of messages from one-minute metrics results per month, per gateway = two metrics x 1,440minutes per day x 30 days = 86,400 messages
Number of messages ingested into cloud per month = 25.92 million messages + 86,400 messages = 26.0064 million messages per monthData StorageSize of message from transforms or metrics = 150 bytes per message
Metrics and transforms data stored in the cloud per month. = Size of message x number of messages per month = 150/1 GB (1024^3 bytes) x 26.0064 M = 3.63 GB per monthBilling SummarySiteWise Edge = One Data Processing Pack x $200 per pack, per month = $200 per monthMessaging = 26 million messages x $1.00 per million messages = $26 per monthData Storage = 1.815 GB* x $0.03 per GB. Month = $0.05 per month*Assuming storage has grown linearly from 0GB to 3.63GB by the end of the month, average storage usage would approximate 1.815 GB.
**Data that remains in storage from prior billing cycles will continue to be metered and billed accordingly.
Total Monthly BillMonthly AWS IoT SiteWise charges = $200 + $26 + $0.05 = $226.05 per month -
Example 7
On top of example 3, you also configure AWS IoT SiteWise to export data from measurements, metrics, transforms, and asset definitions to an Amazon S3 bucket in your AWS account. The AWS IoT SiteWise data export charges are estimated as follows:Usage Summary
Data Export
We will first calculate the number of objects that will be exported for measurements, metrics, and transforms. AWS IoT SiteWise creates a new S3 object for each measurement, metric, or transform that has data in the system every six hours.
Number of objects per measurement, metric, or transform per month = 1 object / 6 hours * 24 hours / 1 day * 30 days / month
Number of objects per measurement, metric, or transform per month = 120 objects / monthTotal number of objects for measurements per month = 120 objects / measurement / month * 1 measurement / equipment * 10 equipment = 1,200 objects per month
Total number of objects for metrics per month = 120 objects / metric / month * 1 metric / equipment * 10 equipment = 1,200 objects per month
Total number of objects for transforms per month = 120 objects / transform / month * 1 transform / measurement * 1 measurement / equipment * 10 equipment = 1,200 objects per month
Total objects per month = 1,200 + 1,200 + 1,200 = 3,600 objects per month
Billing SummaryAdditional Monthly Bill for data export
Data export charges = $0.005 per thousand objects
Total objects exported per month = 3,600 objects
Additional monthly charges for data export = 3,600 objects * $0.005 per thousand objects = $0.018 per month* Additional charges apply for storing exported data in Amazon S3 and for PUT requests made against your Amazon S3 bucket. For more information about these charges, see Amazon S3 pricing.
-
Example 8
On top of example 1, you want to bulk import the last 60 months of historical data for that production line. You will use AWS IoT SiteWise CreateBulkImportJob API to ingest that data from another system to AWS IoT SiteWise.AWS IoT SiteWise will ingest data to the hot and cold storage tiers. Data timestamped after your defined hot tier retention period will be stored in the hot storage tier. Data timestamped prior to your defined retention period will be stored in the cold storage tier. For this pricing calculation we are going to assume that your hot storage tier retention period is 30 days. To learn more about AWS IoT SiteWise data storage, see managing data storage. The AWS IoT SiteWise bulk import charges are estimated as follows:
Usage Summary
Messaging
AWS IoT SiteWise will import 1 month of data to the hot tier storage. Assuming the data for the last 30 days will be imported in 6-hour chunks.
Number of measurements per 6-hour window: 10 devices * 6 hours * 60 min * 60 seconds * 1 measurement per second = 216,000 measurements per 6-hour window.
Assuming 81 bytes per measurement, including overhead, the total volume of data that is ingested every 6 hours is = 216,000 measurements * 81 bytes per measurement / 1024 conversion of byte to KB = 17,086 KB.
Number of messages, measured in 5 KB increments for bulk import = 17,086 / 5 KB per
message = 3,417 messages per 6-hour window.Total monthly messages = 3,417 * 4 (number of 6-hour windows in a day) * 30 = 410,064.
Data Export
Number of objects per measurement, metric, or transform per month = 1 object / 6 hours * 24 hours / 1 day * 30 days / month.
Number of objects per measurement, metric, or transform per month = 120 objects / month.
Total number of objects for measurements per month for 10 equipment = 120 objects /
measurement / month * 1 measurement / equipment * 10 equipment = 1,200 objects per month.AWS IoT SiteWise will export 59 months of data to the cold storage tier.
Total number of objects for 59 months of measurement data = 1,200 objects per month * 59 = 70,800 S3 objects.
Billing Summary
Messaging
Messaging charges = $1.00 per million messages
Total messaging usage for 1-month historical data = 410,064 messages
One time charge for bulk import of 1-month historical data to hot storage tier = 410,064 *
$1.00 per million messages = $0.41
Data Export
Data Export charges = $0.005 per thousand objects
Total objects exported for 59 months = 70,800 objects
One-time charge for bulk import of 59 months historical data to cold storage tier = 70,800*
$0.005 per thousand objects = $0.354
-
Example 9
For a wind farm that is generating 1,000 measurements per second from 100,000 unique data streams, you configure SiteWise to ingest approximately 20% of data streams or tags in near real-time (NRT) and the other 80% to be batched and ingested every several minutes through buffered ingestion. Assume each sensor measurement has 100 bytes. NRT messages contain data points belonging to the same data stream. Buffered messages are constructed from any data stream that has latency tolerance and can be batched. The monthly AWS IoT SiteWise charges are estimated as follows:
Usage Summary
Messaging
Number of measurements ingested in a month = 1,000 * 86,400 seconds / day * 30 days / month = 2,592,000,000 measurements per month
Total number of measurements ingested per month using NRT ingestion = 2,592,000,000 measurements per month * 20% (NRT path) * 100 bytes per measurement / 1 KB increments = 51,840,000 messages per month = 51.84 million messages per month
Total number of data points ingested per month using buffered ingestion = 2,592,000,000 data points per month * 80% (buffered ingestion path) * 100 bytes per measurement / 5 KB increments = 41,472,000 messages per month = 41.472 messages per monthData Processing
Auto-computed aggregates
Number of aggregate computations per message = one computation per message ingested using NRT ingestion
Total number of aggregate computations in hot tier per month = 51,840,000 messages per month = 51.84 million messages per month
Data Storage
Equipment data
Number of measurements per month= 2,592,000,000
Storage per month = 2,592,000,000 measurement per month * 100 bytes / (1024^3 bytes per GB) = 241.40 GB per month
Auto-computed aggregates
SiteWise stores all auto-computed aggregates. The storage used by auto-computed aggregates depends on the number of aggregates and time intervals over which they are calculated. In the example above, six auto-computed aggregates calculated over four time intervals (one-minute, fifteen-minutes, one-hour, one-day) will result in GB per month of data stored.
Storage from auto-computed aggregates = 3,099.63 GB per month*
Total data ingested in NRT per month = 241.40 GB (from Equipment data) + 3,099.63 GB (from Auto-computed aggregates) = 3,341.03 GB per month** This data reflects the data storage volume at the end of the first month.
Billing Summary
Messaging
Total messaging usage in a month = 51.84 million messages (hot tier)
Monthly charges for messaging = 51.840 million * $1.00 per million messages = $51.84 per month
Total messaging usage in a month using buffered ingestion = 41.472 million messages
Monthly charges for messaging to buffered ingestion= 41.472 million * $1 per million messages = $41.47 per month
Data Processing
Data processing charges = $0.50 per million computations
Total computations in a month = 51.84 million computations
Monthly charges for computations = 51.84 million computations * $0.50 per million computations = $25.92 per month
Data Storage
Total data stored at the end of the month = 3,341.03 GB per month
Average storage usage per hour during the month = 3,341.03 GB*
Monthly charges for hot storage = 3,341.03 GB (hot storage) * $0.30 per GB-month = $1,002.31 per month**
Monthly charges for warm storage = 3,341.03 GB * (warm storage) * $0.03 per GB-month = $100.23 per month**
*Assuming storage has grown linearly from 0 GB to 6682.06 GB by the end of the month, average storage usage would approximate 3,341.03 GB.**Data that remains in storage from prior billing cycles will continue to be metered and billed accordingly.
Total Monthly Bill
Monthly AWS IoT SiteWise charges = $51.84 + $41.47 + $25.92 + $1,002.31 + $100.23 = $ 1,221.77 per month
-
Example 10
In addition to data ingestion and storage from example 9, you also add graphs. Using your custom application, you create a dashboard consisting of 10 graphs to monitor your equipment performance. 200 of the 1000 graphical components retrieve data from warm tier showing historical trends, and 800 show near real-time metrics from the hot storage tier. Each graph is pulling the last 100 measurements and refreshes every 15 minutes. For simplicity we are only including the cost of retrieving the data from SiteWise storage.
Usage Summary
Messaging
Near Real-Time Graphs
Number of measurements retrieved per graph in a month= 100 measurements * count of 15- minute intervals in a month = 288,000 measurements per month per graph
Number of messages retrieved from hot storage tier = 100 measurements per interval / 4KB (50 data points) per message * 800 graphs * (count of 15-minute intervals in a month = 4,608,000 messagesAnalytical Graphs
Number of messages retrieved from warm storage tier = 100 measurements per interval / 20KB (250 data points) per interval with min of 10 messages * count of 15-minute intervals in a month * 200 graphs = 5,760,000 messages
Data retrieval per interval amounts to 1 message, however, minimum retrieval from the warm tier is 10 messages per retrieval.Billing Summary
Messaging
Total messaging usage from hot tier in a month = 4,608,000 messages
Total messaging usage from warm tier in a month = 5,760,000 messages
Monthly charges for messaging = (4,608,000 messages from hot + 5,760,000 messages from warm) * $1.00 per million messages = $10.37 per month
Total Monthly Bill
Monthly AWS IoT SiteWise charges = $10.37 per month -
Example 11
In addition to example 2, we now want to recalculate a metric to capture temperature deviations (delta_temp) from two sensors of 10 equipment for 1 year of data (e.g. delta_temp=motor_temp-env_temp). The additional usage and charges are estimated as follows:
Usage Summary
Messaging
Number of metrics per equipment = one metric per equipmentFrequency of messages generated per metric = one message per minute per metric
Number of messages generated in a month = 2 retrievals per metric calculation * 10 equipment * 1440 metrics per day * 30 days = 864,000 messages per month
Additional number of messages retrieved per month = 864,000 messages per month = 0.864 million messages per month
For 12 months: 864,000 messages per month * 12 months = 10.368 million messages
Number of metrics per equipment = one metric per equipment
Frequency of messages generated per metric = one message per minute per metric
Number of messages generated in a month = one message per minute per metric * one metric per equipment * 10 equipment * 1440 minutes per day * 30 days = 432,000 messages per month
Additional number of messages ingested per month = 432,000 messages per month = 0.432 million messages per month
For 12 months: 432,000 messages per month * 12 months = 5.184 million messages
Data Processing
Frequency of computation per metric = one computation per message * one message per second per equipment * one equipment per metric = one computation per second per metric
Number of computations per metric per month = one computation per second per metric * 86,400 seconds per day * 30 days = 2,592,000 computations per metric per month
Number of computations across all equipment per month = 2,592,000 computations per metric per month* one metric per equipment * 10 equipment = 25,920,000 computations = 25.92 million computations per month
For 12 months: 25.92 million computations * 12 months = 311.04 million computationsAuto-computed aggregates
Additional number of aggregate computations per month = one computation per message * 432,000 messages per month = 432,000 computations = 0.432 million computations per month
Additional number of computations per month = 25.92 million (from Custom metrics) + 0.432 million (from Auto-computed aggregates) = 26.352 million computations per month
For 12 months: (0.432 + 26.352) million computations * 12 months = 316.224 million computationsData Storage
Custom metrics
Additional messages ingested per month = 432,000 messages per month
Size of each message = 100 bytes
Additional storage per month = 432,000 messages per month * 100 bytes per message/ (1024^3 bytes per GB) = 0.04 GB per monthFor 12 months: 0.04 GB per month * 12 months = 0.48 GB
Auto-computed aggregates
Additional storage from auto-computed aggregates on custom metrics = 10 equipment, 1 metric, aggregates calculated for 1 min, 5 min, 15 min, 1 hour, 1 day = 0.0000279 GB
Auto computed aggregates calculated for 30 days: 0.0000279 * 30 = 0.000838 GB
Data stored per month = 0.04 GB (for Custom metrics) + 0.00083 GB (from Auto-computed aggregates) = 0.040838 GB per month
Data Stored per year: 0.040838 GB per month * 12 months per year = 0.490 GBTotal Storage = 0.48 GB (custom metrics) + 0.490 GB (aggregates) = 0.97 GB
Billing Summary
Messaging
Messaging usage = 10.368 million messages to retrieve + 5.184 million messages to write resultsCharges for messaging = 15.552 million messages * $1.00 per million messages = $15.55
Data Processing
Additional computations = 316.224 million computationsAdditional charges for computations = 316.224 million computations * $0.50 per million computations = $158.11
Data Storage
Additional data stored = 0.97 GB
Additional charges for storage = 0.97 GB * $0.30 per GB-month = $0.29
Additional Bill for On-Demand Computation
Reprocessing 1 year of data for 5.18 million computations.
Additional bill = $15.55+ $158.11 + $0.40= $174.06 -
Example 12
On top of example 8 you want to import the last 60 months of historical data. You will use AWS IoT SiteWise CreateBulkImportJob API to ingest data from another system to AWS IoT SiteWise.
AWS IoT SiteWise will ingest data to the hot, warm, and cold storage tiers. For this pricing calculation, we are going to assume that your hot storage tier retention period is 30 days and warm tier retention period is 5 years. The AWS IoT SiteWise import charges are estimated as follows:
Usage Summary
1 – Ingesting to hot storage tier
Messaging
Total number of measurements ingested per month using buffered ingestion.
2,592,000,000 measurements per month * 100% (buffered ingestion) * 100 bytes per measurement / 5 KB increments = 50,625,000 messages per month
Data Processing
Number of aggregate computations per month = one computation per message
50,625,000 messages per month = 50,625,000 computations per monthData Storage
Size of each measurement = 100 bytes per data points (assuming each measurement ingested is 100 bytes)
Monthly storage charges = 2,592,000,000 measurement per month * 100 bytes / (1024^3 bytes per GB) = 241.4 GB per monthAuto-computed aggregates
Storage from auto-computed aggregates = 3,099.63 GB per month
Total data stored per month in hot storage = 241.40 GB (from Equipment data) + 3,099.63 GB (from Auto-computed aggregates) = 3,341.03 GB per month
2- Ingesting to warm storage tier
Messaging
50,625,000 Messages per month * 60 months in warm storage = 3,037,500,000 messagesData Storage
Equipment data
241.40 GB per month * 60 months in warm storage = 14,484 GB total for equipment data
Auto-computed aggregates
3,099.63 GB per month * 60 months in warm storage =185,977.8 GB total for equipment dataTotal data stored per month in warm storage = 14,484GB + 185,977.8GB = 200,461.8 GB
Billing Summary
Messaging
One-time charge for import of 60-month historical data to warm storage tier = 3,037.500 million messages
One-time charge for Messaging of 60 months of historical data = $3,037.50
Data Processing
Pre-Computed aggregate computations are free for buffered historical data ingestion.Data Storage
Storage of 60 months of historical data to warm storage tier = 3,341.03 GB per month * 60 months * $0.03 per GB = $ 6,013.85Total Messaging + Data Storage Cost = $3,037.50 + $6,013.85 = $9,051.35
-
Example 13
You want to run analytical queries on the last 60 months of data for 30 pieces of equipment. Each equipment has 50 sensors. You will use AWS IoT SiteWise ExecuteQuery API. For this pricing calculation we are going to assume that your hot storage tier retention period is 30 days and warm tier retention period is unbounded (unlimited).
Usage Summary
Messaging
Data retrieval from hot storage tier per asset = 23,063 measurementsNumber of messages retrieved from Hot storage per asset = 23,063 measurements * 100 byte per measurement/ ( 4 KB or 50 measurements) = 462 messages
Number of messages retrieved from Hot storage for 30 assets = 462 messages * 30 assets = 13,860 messages
Data retrieval from warm storage tier per asset = 23,063 measurements * 59 months = 1,360,717 measurements
Number of messages retrieved from Warm storage per asset = 1,360,717 measurements / (20 KB or 250 measurements) = 5,443 messages
Number of messages retrieved from Warm storage for 30 asset = 5,443 messages * 30 equipment = 163,290 messages
Query Requests
Assuming total Measurements to scan per asset (Data scan): 1,383,772Measurements to scan per month per asset : 1,383,772 / 60 months = 23,063 measurements
per monthData Scan Volume: 30 asset * 1,383,772 data scanned per asset = 41,513,160 measurements
Data Scan in GB = 3.866Billing Summary
Messaging
From hot tier: 13,838 messages at $1 per million = $0.013838Warm tier: 163,290 messages at $1 per million = $0.16329
Total Messaging Cost = $0.17712
Query Requests
$0.20 per GB scanned increments of 100KBSQL cost per data scanned = 3.867 GB * $0.20 per GB Scanned = $0.7732
Total Messaging + SQL data scanned Cost = $0.17712 + $0.7732= $0.95
Refer to the AWS IoT SiteWise user guide, API reference and application guide for more information.