Exam DP-203 topic 2 question 33 discussion

You do not need a Window function. You just process the data and perform the geospatial check as it arrives. See the same example here: https://docs.microsoft.com/en-us/azure/stream-analytics/geospatial-scenarios

1. Azure Stream Analytics 2. No Window 3. Point within Polygon

Got this question in exam 22/06/2024. but only 3 options in Window without tumbling

I see that people are mentioning tumbling window here for this question. Would the query look something like this? SELECT Collect() as allEvents INTO Output FROM DeviceStreamInput JOIN GeoFenceReference ON ST_WITHIN(DeviceStreamInput.GeoPosition, SiteReferenceInput.Geofence) = 0 WHERE DeviceStreamInput.DeviceID = GeoFenceReference.DeviceID GROUP BY TumblingWindow(second, 30)

1. Azure Stream Analytics 2. Sliding (change on event) or No window for me 3. Point within Polygon

Wrong, chatgpt : - **Service**: Azure Stream Analytics – This service is ideal for processing real-time streaming data from IoT devices. - **Window**: No window – As you're processing each GPS event as it arrives, you don't need to aggregate over a time window but process each event individually to check if it's within the expected area. - **Analysis type**: Point within polygon – This analysis type is used for geospatial analytics, where you're checking if a point (the GPS position) is within a predefined polygon (the expected geographical area for the vehicle).

It is clear tumbling window guys. requirements:- 1) new data created in every 1 min 2) if any out of range problem is there then need to send with in 30 sec. so window size is 1 min, we will have 1 min of data., so we will not miss data. hop size will be 30 sec, so we can process the data in 30 sec and report the problem if any.

Correct answers are: 1. Azure Stream analytics 2. Tumbling window 3. Point within Polygon

It sounds like you want to use Azure Stream Analytics for this task. Stream Analytics is a real-time analytics service that allows you to analyze and process high volumes of streaming data from various sources, such as Azure Event Hubs. For the window, you should use a Tumbling window. A tumbling window is a fixed-sized, non-overlapping window of data. It is well-suited for this scenario because you want to process the data once per minute, and a tumbling window with a size of 1 minute would allow you to do this. For the analysis type, you should use Point within polygon. This analysis type allows you to determine whether a GPS position falls within a specific geographical area. You can use the CSV file in the Azure Data Lake Storage Gen2 container to define the expected geographical areas for each vehicle.

Choosing no window is totally wrong. No window means you have to process a msg everytime a event happened. It's costly. But a session window can be trigged when there is a event and can combine all the events in maxduration size(here is 30s) to add a packet of events at one time. And other types of windows are not suitable for the situation here.

I think the answer is correct, should be hopping window with window size is 30 seconds and hope size is 1 minute.

Answer is correct, agree to the point made by VyshakhUnnikrishnan

when a GPS position is outside the expected area, a message is added to another event hub for processing within 30 seconds. - this has to be taken in to account No Window

1. Azure Stream Analytics 2. No Window 3. Point within Polygon

i thing it is correct

I am not sure about tumbling window here. I think it would work but what would be the requirement to send an event out every thirty seconds with no data i.e no vehicle has traveled outside the geographic area. In my opinion, a no window could send data to another event hub as needed.

It's Tumbling. No Window: it should run 500 times per minute (500 messages are sent per minute) -> costly. Tumbling Window: if we configure 15s -> it runs 4 times per minute -> much better. Hopping Window: Some messages are processed twice (don't care allowing duplication or not) -> costly.