IoT Time Series Bulk Service¶
The IoT Time Series Bulk Service is a variant to the IoT Time Series Service. It is designed for importing big amounts of data into MindSphere in bulk and can handle both high frequency simulation data and performance data.
For accessing this service you need to have the respective roles listed in IoT Time Series Service roles and scopes.
In contrast to the IoT Time Series Service, this service ingests data in batch mode and can return more than 2,000 items per response when reading the stored data. It returns as many items as possible until size and time limitations on the MindSphere Gateway are exceeded. If it cannot return all requested data in a single response, a link for requesting the next data set is added to the response body. The URL of this link is based on the already returned data and the original request: The start time is set to the timestamp of the last returned record and the maximum number of records is reduced by the number of already returned records.
The IoT Time Series Bulk Service accepts up to 100 consecutive files per import job. For simulation asset, the hour part of all timestamps within these files must be equal. Multiple import requests for the same hour can be handled to allow importing files with overlapping time periods. For performance asset, all the timestamps should belong to the same day. For Bulk Ingest of time series, only time stamps in ISO format are supported.
The import requests are processed sequentially with data for the following assets:
- simulation asset, for the same hour
- performance asset, for the same day
Previously stored time series data is overwritten when uploading new data for the same time period.
The IoT Time Series Bulk Service exposes its API for realizing the following tasks:
- Import high frequency time series data for simulation assets
- Import time series data for performance assets in bulk
- Read time series data
- Currently, only the IoT File Service is supported as data source.
- The maximum volume of time series data for one hour cannot exceed 350 MB for simulation asset.
- The maximum volume of time series data for one day cannot exceed 1 GB for performance asset.
- Import jobs for performance assets are allowed for data older than 30 minutes only.
- Import jobs for performance assets are rejected if timestamps with higher precision than milliseconds are used.
- The '+' character used to include timezone in timestamp must be encoded to '%2B' in time range. For example, 2017-09-21T14:08:00%2B02:00 instead of 2017-09-21T14:08:00+02:00.
A company is already collecting data for an existing field device. They want to correlate this with simulation data created during the engineering or testing process in order to derive relevant KPIs. The company uploads the simulation data using the IoT File Service and triggers the import with the IoT Time Series Bulk Service. After the importing the data can be consumed via this service to be used by an application that also has access to the Time Series data of the field device.
Any questions left?
Except where otherwise noted, content on this site is licensed under the MindSphere Development License Agreement.