File paths and schemas of data saved in the cold tier - Amazon IoT SiteWise
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China.

File paths and schemas of data saved in the cold tier

Amazon IoT SiteWise stores your data in the cold tier by replicating time series, including measurements, metrics, transforms and aggregates, and also asset and asset model definitions. The following describes the file paths and schemas of data that is sent to the cold tier.

Equipment data (measurements)

Amazon IoT SiteWise exports equipment data (measurements) to the cold tier once every six hours. Raw data is saved in the cold tier in the Apache AVRO (.avro) format.

File path

Amazon IoT SiteWise stores equipment data (measurements) in the cold tier using the following template.

{keyPrefix}/raw/startYear={startYear}/startMonth={startMonth}/startDay={startDay}/seriesBucket={seriesBucket}/raw_{timeseriesId}_{startTimestamp}_{quality}.avro

Every file path to raw data in Amazon S3 contains the following components.

Path component Description

keyPrefix

The Amazon S3 prefix that you specified in the Amazon IoT SiteWise storage configuration. Amazon S3 uses the prefix as a folder name in the bucket.

raw

The folder that stores time series data from equipment (measurements). The raw folder is saved in the prefix folder.

seriesBucket

A hexadecimal number between 00 and ff. This number is derived from timeSeriesId. This partition is used to increase throughput when Amazon IoT SiteWise writes to the cold tier. When you use Amazon Athena to run queries, you can use the partition for fine-grain partitioning to improve query performance.

seriesBucket and timeSeriesBucket in the asset metadata are the same number.

startYear

The year of the exclusive start time associated with the time series data.

startMonth

The month of the exclusive start time associated with the time series data.

startDay

The day of the month of the exclusive start time associated with the time series data.

fileName

The file name uses the underscore (_) character as a delimiter to separate the following:

  • The raw prefix.

  • The timeSeriesId value.

  • The epoch timestamp of the exclusive start time associated with the time series data.

  • The quality of the data. Valid values: GOOD, BAD, and UNCERTAIN. For more information, see AssetPropertyValue in the Amazon IoT SiteWise API Reference.

The file is saved in the .avro format by using the Snappy compression.

Example file path to raw data in the cold tier

keyPrefix/raw/startYear=2021/startMonth=1/startDay=2/seriesBucket=a2/raw_7020c8e2-e6db-40fa-9845-ed0dddd4c77d_95e63da7-d34e-43e1-bc6f-1b490154b07a_1609577700_GOOD.avro

Fields

The schema of raw data that is exported to the cold tier contains the following fields.

Field name Supported types Default type Description

seriesId

string

N/A

The ID that identifies the time series data from equipment (measurements). You can use this field to join raw data and asset metadata in queries.

timeInSeconds

long

N/A

The timestamp date, in seconds, in the Unix epoch format. Fractional nanosecond data is provided by offsetInNanos.

offsetInNanos

long

N/A

The nanosecond offset from timeInSeconds.

quality

string

N/A

The quality of the time series value.

doubleValue

double or null

null

Time series data of type double (floating point number).

stringValue

string or null

null

Time series data of type string (sequence of characters).

integerValue

int or null

null

Time series data of type integer (whole number).

booleanValue

boolean or null

null

Time series data of type Boolean (true or false).

jsonValue

string or null

null

Time series data of type JSON (complex data types stored as a string).

recordVersion

long or null

null

The version number for the record. You can use the version number to select the latest record. Newer records have larger version numbers.

Example raw data in the cold tier

{"seriesId":"e9687d2a-0dbe-4f65-9ed6-6f443cba41f7_95e63da7-d34e-43e1-bc6f-1b490154b07a","timeInSeconds":1625675887,"offsetInNanos":0,"quality":"GOOD","doubleValue":{"double":0.75},"stringValue":null,"integerValue":null,"booleanValue":null,"jsonValue":null,"recordVersion":null} {"seriesId":"e9687d2a-0dbe-4f65-9ed6-6f443cba41f7_95e63da7-d34e-43e1-bc6f-1b490154b07a","timeInSeconds":1625675889,"offsetInNanos":0,"quality":"GOOD","doubleValue":{"double":0.69},"stringValue":null,"integerValue":null,"booleanValue":null,"jsonValue":null,"recordVersion":null} {"seriesId":"e9687d2a-0dbe-4f65-9ed6-6f443cba41f7_95e63da7-d34e-43e1-bc6f-1b490154b07a","timeInSeconds":1625675890,"offsetInNanos":0,"quality":"GOOD","doubleValue":{"double":0.66},"stringValue":null,"integerValue":null,"booleanValue":null,"jsonValue":null,"recordVersion":null} {"seriesId":"e9687d2a-0dbe-4f65-9ed6-6f443cba41f7_95e63da7-d34e-43e1-bc6f-1b490154b07a","timeInSeconds":1625675891,"offsetInNanos":0,"quality":"GOOD","doubleValue":{"double":0.92},"stringValue":null,"integerValue":null,"booleanValue":null,"jsonValue":null,"recordVersion":null} {"seriesId":"e9687d2a-0dbe-4f65-9ed6-6f443cba41f7_95e63da7-d34e-43e1-bc6f-1b490154b07a","timeInSeconds":1625675892,"offsetInNanos":0,"quality":"GOOD","doubleValue":{"double":0.73},"stringValue":null,"integerValue":null,"booleanValue":null,"jsonValue":null,"recordVersion":null}

Metrics, transforms, and aggregates

Amazon IoT SiteWise exports metrics, transforms, and aggregates to the cold tier once every six hours. Metrics, transforms, and aggregates are saved in the cold tier in the Apache AVRO (.avro) format.

File path

Amazon IoT SiteWise stores metrics, transforms, and aggregates in the cold tier using the following template.

{keyPrefix}/agg/startYear={startYear}/startMonth={startMonth}/startDay={startDay}/seriesBucket={seriesBucket}/agg_{timeseriesId}_{startTimestamp}_{quality}.avro

Every file path to metrics, transforms, and aggregates in Amazon S3 contains the following components.

Path component Description

keyPrefix

The Amazon S3 prefix that you specified in the Amazon IoT SiteWise storage configuration. Amazon S3 uses the prefix as a folder name in the bucket.

agg

The folder that stores time series data from metrics. The agg folder is saved in the prefix folder.

seriesBucket

A hexadecimal number between 00 and ff. This number is derived from timeSeriesId. This partition is used to increase throughput when Amazon IoT SiteWise writes to the cold tier. When you use Amazon Athena to run queries, you can use the partition for fine-grain partitioning to improve query performance.

seriesBucket and timeSeriesBucket in the asset metadata are the same number.

startYear

The year of the exclusive start time associated with the time series data.

startMonth

The month of the exclusive start time associated with the time series data.

startDay

The day of the month of the exclusive start time associated with the time series data.

fileName

The file name uses the underscore (_) character as a delimiter to separate the following:

  • The raw prefix.

  • The timeSeriesId value.

  • The epoch timestamp of the exclusive start time associated with the time series data.

  • The quality of the data. Valid values: GOOD, BAD, and UNCERTAIN. For more information, see AssetPropertyValue in the Amazon IoT SiteWise API Reference.

The file is saved in the .avro format by using the Snappy compression.

Example file path to metrics in the cold tier

keyPrefix/agg/startYear=2021/startMonth=1/startDay=2/seriesBucket=a2/agg_7020c8e2-e6db-40fa-9845-ed0dddd4c77d_95e63da7-d34e-43e1-bc6f-1b490154b07a_1609577700_GOOD.avro

Fields

The schema of metrics, transforms, and aggregates that are exported to the cold tier contains the following fields.

Field name Supported types Default type Description

seriesId

string

N/A

The ID that identifies the time series data from equipment, metrics, or transforms. You can use this field to join raw data and asset metadata in queries.

timeInSeconds

long

N/A

The timestamp date, in seconds, in the Unix epoch format. Fractional nanosecond data is provided by offsetInNanos.

offsetInNanos

long

N/A

The nanosecond offset from timeInSeconds.

quality

string

N/A

The quality by which to filter asset data.

resolution

string

N/A

The time interval over which to aggregate data.

count

double or null

null

The total number of data points for the given variables over the current time interval.

average

double or null

null

The mean of the given variables' values over the current time interval.

min

double or null

null

The minimum of the given variables' values over the current time interval.

max

boolean or null

null

The maximum of the given variables' values over the current time interval.

sum

string or null

null

The sum of the given variables' values over the current time interval.

recordVersion

long or null

null

The version number for the record. You can use the version number to select the latest record. Newer records have larger version numbers.

Example Metric data in the cold tier

{"seriesId":"f74c2828-5317-4df3-ba16-6d41b5bcb531","timeInSeconds":1637334060,"offsetInNanos":0,"quality":"GOOD","resolution":"PT1M","count":31.0,"average":{"double":16.0},"min":{"double":1.0},"max":{"double":31.0},"sum":{"double":496.0},"recordVersion":null} {"seriesId":"f74c2828-5317-4df3-ba16-6d41b5bcb531","timeInSeconds":1637334120,"offsetInNanos":0,"quality":"GOOD","resolution":"PT1M","count":29.0,"average":{"double":46.0},"min":{"double":32.0},"max":{"double":60.0},"sum":{"double":1334.0},"recordVersion":null} {"seriesId":"f74c2828-5317-4df3-ba16-6d41b5bcb531","timeInSeconds":1637334540,"offsetInNanos":0,"quality":"GOOD","resolution":"PT1M","count":31.0,"average":{"double":16.0},"min":{"double":1.0},"max":{"double":31.0},"sum":{"double":496.0},"recordVersion":null} {"seriesId":"f74c2828-5317-4df3-ba16-6d41b5bcb531","timeInSeconds":1637334600,"offsetInNanos":0,"quality":"GOOD","resolution":"PT1M","count":29.0,"average":{"double":46.0},"min":{"double":32.0},"max":{"double":60.0},"sum":{"double":1334.0},"recordVersion":null} {"seriesId":"f74c2828-5317-4df3-ba16-6d41b5bcb531","timeInSeconds":1637335020,"offsetInNanos":0,"quality":"GOOD","resolution":"PT1M","count":31.0,"average":{"double":16.0},"min":{"double":1.0},"max":{"double":31.0},"sum":{"double":496.0},"recordVersion":null}

Asset metadata

When you enable Amazon IoT SiteWise to export data to the cold tier for the first time, asset metadata is exported to the cold tier. After the initial configuration, Amazon IoT SiteWise exports asset metadata to the tier only when you change asset model definitions or asset definitions. Asset metadata is saved in the cold tier in the Newline Delimited JSON (.ndjson) format.

File path

Amazon IoT SiteWise stores asset metadata in the cold tier using the following template.

{keyPrefix}/asset_metadata/asset_{assetId}.ndjson

Every file path to asset metadata in the cold tier contains the following components.

Path component Description

keyPrefix

The Amazon S3 prefix that you specified in the Amazon IoT SiteWises storage configuration. Amazon S3 uses the prefix as a folder name in the bucket.

asset_metadata

The folder that stores asset metadata. The asset_metadata folder is saved in the prefix folder.

fileName

The file name uses the underscore (_) character as a delimiter to separate the following:

  • The asset prefix.

  • The assetId value.

The file is saved in the .ndjson format.

Example file path to asset metadata in the colder tier

keyPrefix/asset_metadata/asset_35901915-d476-4dca-8637-d9ed4df939ed.ndjson

Fields

The schema of asset metadata that is exported to the cold tier contains the following fields.

Field name Description

assetId

The ID of the asset.

assetName

The name of the asset.

assetModelId

The ID of the asset model used to create this asset.

assetModelName

The name of the asset model.

assetPropertyId

The ID of the asset property.

assetPropertyName

The name of the asset property.

assetPropertyDataType

The data type of the asset property.

assetPropertyUnit

The unit of the asset property (for example, Newtons and RPM).

assetPropertyAlias

The alias that identifies the asset property, such as an OPC-UA server data stream path (for example, /company/windfarm/3/turbine/7/temperature).

timeSeriesId

The ID that identifies the time series data from equipment, metrics, or transforms. You can use this field to join raw data and asset metadata in queries.

timeSeriesBucket

A hexadecimal number between 00 and ff. This number is derived from timeSeriesId. This partition is used to increase throughput when Amazon IoT SiteWise writes to the cold tier. When you use Amazon Athena to run queries, you can use the partition for fine-grain partitioning to improve query performance.

timeSeriesBucket and seriesBucket in the file path to raw data are the same number.

assetCompositeModelDescription

The description of the composite model.

assetCompositeModelName

The name of the composite model.

assetCompositeModelType

The type of the composite model. For alarm composite models, this type is AWS/ALARM.

assetCreationDate

The date the asset was created, in Unix epoch time.

assetLastUpdateDate

The date the asset was last updated, in Unix epoch time.

assetStatusErrorCode

The error code.

assetStatusErrorMessage

The error message.

assetStatusState

The current status of the asset.

Example asset metadata in the cold tier

{"assetId":"7020c8e2-e6db-40fa-9845-ed0dddd4c77d","assetName":"Wind Turbine Asset 2","assetModelId":"ec1d924f-f07d-444f-b072-e2994c165d35","assetModelName":"Wind Turbine Asset Model","assetPropertyId":"95e63da7-d34e-43e1-bc6f-1b490154b07a","assetPropertyName":"Temperature","assetPropertyDataType":"DOUBLE","assetPropertyUnit":"Celsius","assetPropertyAlias":"USA/Washington/Seattle/WT2/temp","timeSeriesId":"7020c8e2-e6db-40fa-9845-ed0dddd4c77d_95e63da7-d34e-43e1-bc6f-1b490154b07a","timeSeriesBucket":"f6","assetArn":null,"assetCompositeModelDescription":null,"assetCompositeModelName":null,"assetCompositeModelType":null,"assetCreationDate":1619466323,"assetLastUpdateDate":1623859856,"assetStatusErrorCode":null,"assetStatusErrorMessage":null,"assetStatusState":"ACTIVE"} {"assetId":"7020c8e2-e6db-40fa-9845-ed0dddd4c77d","assetName":"Wind Turbine Asset 2","assetModelId":"ec1d924f-f07d-444f-b072-e2994c165d35","assetModelName":"Wind Turbine Asset Model","assetPropertyId":"c706d54d-4c11-42dc-9a01-63662fc697b4","assetPropertyName":"Pressure","assetPropertyDataType":"DOUBLE","assetPropertyUnit":"KiloPascal","assetPropertyAlias":"USA/Washington/Seattle/WT2/pressure","timeSeriesId":"7020c8e2-e6db-40fa-9845-ed0dddd4c77d_c706d54d-4c11-42dc-9a01-63662fc697b4","timeSeriesBucket":"1e","assetArn":null,"assetCompositeModelDescription":null,"assetCompositeModelName":null,"assetCompositeModelType":null,"assetCreationDate":1619466323,"assetLastUpdateDate":1623859856,"assetStatusErrorCode":null,"assetStatusErrorMessage":null,"assetStatusState":"ACTIVE"} {"assetId":"7020c8e2-e6db-40fa-9845-ed0dddd4c77d","assetName":"Wind Turbine Asset 2","assetModelId":"ec1d924f-f07d-444f-b072-e2994c165d35","assetModelName":"Wind Turbine Asset Model","assetPropertyId":"8cf1162f-dead-4fbe-b468-c8e24cde9f50","assetPropertyName":"Max Temperature","assetPropertyDataType":"DOUBLE","assetPropertyUnit":null,"assetPropertyAlias":null,"timeSeriesId":"7020c8e2-e6db-40fa-9845-ed0dddd4c77d_8cf1162f-dead-4fbe-b468-c8e24cde9f50","timeSeriesBucket":"d7","assetArn":null,"assetCompositeModelDescription":null,"assetCompositeModelName":null,"assetCompositeModelType":null,"assetCreationDate":1619466323,"assetLastUpdateDate":1623859856,"assetStatusErrorCode":null,"assetStatusErrorMessage":null,"assetStatusState":"ACTIVE"}

Asset hierarchy metadata

When you enable Amazon IoT SiteWise to save data the in cold tier for the first time, asset hierarchy metadata is exported to the cold tier. After the initial configuration, Amazon IoT SiteWise exports asset hierarchy metadata to the cold tier only when you make changes to asset model or asset definitions. Asset hierarchy metadata is saved in the cold tier in the Newline Delimited JSON (.ndjson) format.

File path

Amazon IoT SiteWise stores asset hierarchy metadata in the cold tier using the following template.

{keyPrefix}/asset_hierarchy_metadata/{parentAssetId}_{hierarchyId}.ndjson

Every file path to asset hierarchy metadata in the cold tier contains the following components.

Path component Description

keyPrefix

The Amazon S3 prefix that you specified in the Amazon IoT SiteWise storage configuration. Amazon S3 uses the prefix as a folder name in the bucket.

asset_hierarchy_metadata

The folder that stores asset hierarchy metadata. The asset_hierarchy_metadata folder is saved in the prefix folder.

fileName

The file name uses the underscore (_) character as a delimiter to separate the following:

  • The parentAssetId value.

  • The hierarchyId value.

The file is saved in the .ndjson format.

Example file path to asset hierarchy metadata in the cold tier

keyPrefix/asset_hierarchy_metadata/35901915-d476-4dca-8637-d9ed4df939ed_c5b3ced8-589a-48c7-9998-cdccfc9747a0.ndjson

Fields

The schema of asset hierarchy metadata that is exported to the cold tier contains the following fields.

Field name Description

sourceAssetId

The ID of the source asset in this asset relationship.

targetAssetId

The ID of the target asset in this asset relationship.

hierarchyId

The ID of the hierarchy.

associationType

The association type of this asset relationship.

The value must be CHILD. The target asset is a child asset of the source asset.

Example asset hierarchy metadata in the cold tier

{"sourceAssetId":"80388e72-2284-44fb-9c89-bfbaf0dfedd2","targetAssetId":"2b866c25-0c74-4750-bdf5-b73683c8a2a2","hierarchyId":"bbed9f59-0412-4585-a61d-6044db526aee","associationType":"CHILD"} {"sourceAssetId":"80388e72-2284-44fb-9c89-bfbaf0dfedd2","targetAssetId":"6b51246e-984d-460d-bc0b-470ea47d1e31","hierarchyId":"bbed9f59-0412-4585-a61d-6044db526aee","associationType":"CHILD"}

To view your data in the cold tier

  1. Navigate to the Amazon S3 console.

  2. In the navigation pane, choose Buckets, and then choose your Amazon S3 bucket.

  3. Navigate to the folder that contains the raw data, asset metadata, or asset hierarchy metadata.

  4. Select the files, and then from Actions, choose Download.

Storage data index files

Amazon IoT SiteWise uses these files to optimize data query performance. They appear in your Amazon S3 bucket, but you don't need to use them.

File path

Amazon IoT SiteWise stores data index files in the cold tier using the following template.

keyPrefix/index/series=timeseriesId/startYear=startYear/startMonth=startMonth/startDay=startDay/index_timeseriesId_startTimestamp_quality

Example file path to data storage index file

keyPrefix/index/series=7020c8e2-e6db-40fa-9845-ed0dddd4c77d_95e63da7-d34e-43e1-bc6f-1b490154b07a/startYear=2022/startMonth=02/startDay=03/index_7020c8e2-e6db-40fa-9845-ed0dddd4c77d_95e63da7-d34e-43e1-bc6f-1b490154b07a_1643846400_GOOD