Use Case
Let us consider some weather stations which are connected to Senseforce via the SF Edge client. They regularly send data about the weather like temperature, wind speed, humidity, etc. However, it doesn't send any metadata which could tell us information such as the station type, location, date of installation, etc. We have this information in the table below:
Weather Station | Location | Type | Installed |
WS1 | Dornbirn | Medium | 2015-03-01 |
WS2 | Feldkirch | Small | 2014-03-01 |
WS3 | Imst | Medium | 2016-03-01 |
In its current format, this information is not very useful. If we had a link between this information and the data from the stations, we could do things like filter out results from specific station types, or aggregate all the data from a specific location. This is where Machine Master Data comes in.
We start by creating a dimension called "Location", which will store details like the station's latitude and longitude. We could store these details directly in the "Weather Station" dimension that we are going to create next, however, this type of information sounds like it will be useful for other dimensions as well.
Now that we have our Location dimension, we create our "Weather Station" dimension using "Location" as a subdimension.
We can now create the instances for our weather stations. We fill out the fields in the Instance editor, clicking the "Location" field to reveal its sub-fields. If we have already created the Instance for the location we want to set, we could use the dropdown on the right side of the "Location" field to use this existing value.
After we have repeated this process for the other two weather stations, we can apply the instances to their corresponding things. In the clip below, WS1 is applied to the thing "MqttNetPerformanceTest15".
Before we can see our changes in the Dataset Editor, we need to publish our machine master data.
Now we can use the extra information we have about our weather stations in our datasets!
Last modified 1yr ago