Modern Data Warehousing ​Hackathon

Challenge 1: Select and provision storage for an enterprise data lake 

In this challenge, you will establish an enterprise data lake. 

Learning objectives: 

• Compare and contrast Azure storage offerings 
• Provision the selected Azure storage service 

Challenge 2: Ingest data from cloud sources 

In this challenge, you will extract the initial Southridge data from Azure SQL databases and a Cosmos DB collection. 

Learning objectives: 

• Orchestrate the ingestion of data from multiple cloud-based sources to a single cloud-based store 
• Ensure the protection of specific customer data at all times leveraging the current technology set and solution architecture 

Challenge 3: Pull data from on-premises and establish source control 

In this challenge, you will incorporate additional data sources into a new data lake. While the initial data was extracted from cloud based Azure SQL databases, the data from this challenge comes from on-premises data stores. 

Learning objectives: 

• Orchestrate the ingestion of data specifically from maintained “on-premises” solutions 
• Implement a cloud-based source control repository for the developed solution 

Challenge 4: Transform and normalize data within the lake and establish branch policies 

In this challenge, you will find (if you have not already!) that the data from these sources has a variety of data types and formats; it is up to the team to preprocess it to be consistent for downstream consumers. 

Learning objectives: 

• Transform data into a normalized schema for downstream consumption 

• Create new policies to make certain all future changes leverage an appropriate review process 

Challenge 5: Populate a data warehouse and implement unit tests 

In this challenge, you will need to fulfill the reporting needs of the business! A Power BI report will be provided to the team, but it’s up to them to create and populate the star schema to which it connects. The definition of the target schema is also supplied. From a DevOps perspective, the team will need to establish their approach to unit testing. 

Learning objectives: 

• Transform the data from the various source systems into a common data warehouse schema to support the generation of specific reports mandated by the business 
• Orchestrate the dataflow into the data warehouse in an automated manner 
• Build out unit tests across core components of the data pipeline 
• Integrate automated testing into the code review process 

Challenge 6: Differential data loads and telemetry 

In this challenge, you will need to address the ongoing needs of the business. Data from each new day of business needs to be added to the data lake, but it would be inefficient to repeatedly process all the historical data. You will need to implement an incremental load and establish a logging and telemetry solution by which you can monitor the amount of newly incorporated data. 

Learning objectives: 

• Modify the solution to include doing differential data loads as well as the original bulk load 
• Automate data load and processing to run daily 

• Implement rich telemetry into the dataflow and deployment pipelines 
• Add error handling to raise pipeline issues in real-time 

Challenge 7: Automated deployment with validation and approval 

In this challenge, you will need to automate gated deployments of new and updated solutions. 

Learning objectives: 

• Operationalize the solution deployment process through automation 
• Create and implement a testing environment 
• Implement automated deployment processes and policies