Exam DP-201 topic 7 question 1 discussion

should be A & C as the final result should result in synapse

I believe daily inventory data are going through ADLK Gen2 (As needed for staging) and go to to Analytical Data Store (Synapse). Another thing is "Daily inventory data comes from a Microsoft SQL server located on a private network". I believe this is on prem server. They didn't mentioned how this sql server consume the daily inventory data. Previous day sales data which are already in Synapse (Assumption is daily inventory data migrated from sql server to datalake and then to Synapse). Current Day's sales data can be ingested to Synapse directly through event hub by using Azure Stream Analytics. OR current day's sales data can be store to blob storage from Event Hub. You can use ADF to load those files to Synapse every 10 min. Then after you can create direct Power BI dashboard which can run T-SQL to calculate inventory level in Synapse and feed it to Power BI report. This can be live report/dashboard which can be share to every store. Answer should be A and E.

Some outputs types support partitioning, and output batch sizes vary to optimize throughput. The following table shows features that are supported for each output type: TABLE 1 Output type Partitioning Sehttps://docs.microsoft.com/en-us/azure/stream-analytics/stream-analytics-define-outputs ASA cannot output to Databricks Azure Data Lake Storage Gen 1 Yes Azure Active Directory user , Managed Identity Azure SQL Database Yes, optional. SQL user auth, Managed Identity (preview) Azure Synapse Analytics Yes SQL user auth, Managed Identity (preview) Blob storage and Azure Data Lake Gen 2 Yes Access key, Managed Identity (preview) Azure Event Hubs Yes, need to set the partition key column in output configuration. Access key, Managed Identity (preview) Power BI No Azure Active Directory user, Managed Identity Azure Table storage Yes Account key Azure Service Bus queues Yes Access key Azure Service Bus topics Yes Access key Azure Cosmos DB Yes Access key Azure Functions Yes Access key

Previous day's data will be in the DW so better to do the calculation on the DW. Use Transact-SQL to calculate the inventory levels.

I will not go for databricks because it is mentioned expressroute or VPN will not be there. https://docs.microsoft.com/en-us/azure/databricks/administration-guide/cloud-configurations/azure/on-prem-network

E- using AVRO not useful as polybase does not support AVRO So only left option is A and C

"Stage inventory data in Azure Data Lake Storage Gen2 before loading the data into the analytical data store" D is out PolyBase does not support Avro B is out Running DataFactory every 10 minutes is unncessary E is out A and C are therefore correct

D is obviously not an option since it does not fulfil the requirement of calculating quickly! Option C should be selected instead - it receives the last day's inventory data from the Data lake which is mentioned in the case study (it is used for staging the data before it is loaded into the Data Warehouse.) -> Answer: A & C

If it is A and E , then aggregation by store and product is done twice, which may not give the right result. So , from a business perspective, A and C makes more sense.

B: is incorrect because Avro files doesn't support Polybase. C: is incorrect because first of all I am not sure you can output to result directly to Azure databrick. Even if you can then again you need to get previous sales data from datalake or from synapse. Then calculate the inventory level by comparing both of them and send it to Synapse and feed the Power BI report. This doesn't make sense. In Synapse, based on right distribution key on large table, your join query works faster. I believe it will be fast and cheaper if you calculate the inventory level in Synapse compare to databrick. D: is incorrect because you don't want to store result in blob storage. It just doesn't make sense. With Synapse and Power BI, you can see historical inventory level by day, by store, by product etc if you configure it right way.

Answer is A n E only as in the question it has been stated that "Stage inventory data in Azure Data Lake Storage Gen2 before loading the data into the analytical data store" , so the final output should go to analytical data store and azure storage should be used as staging which in question been told to remove the files from it as soon as data is uploaded into the final stage.

E can not be correct, because the use of an time scheduled trigger instead of an event blob-trigger. Update have to be done as close to real-time as possible!!

Clearly the assumption is that the inventory of the day before is already in Synapse Analytics. On that basis the answer is correct: in A you aggregate the data, then load them into Synapse and there you use T-SQL statement to calculate the difference with the day before. In E you copy in Synapse the raw data, then use T-SQL to aggregate them and more T-SQL to calculate the difference with the day before

A, B seems to be correct as well. with 10 minutes schedule using adf as mentioned in E option, there is a delay.

Option E does not calculate the inventory. Only groups input by store and prodcut. Databricks can calculate the inventory by reading eventhub data and inventory data from data lake (staged) or synapse. Since this is neither mentioned nor left out of option C. I would suggest options A and C to be the correct answers.

Why not A, C?