Explanation

In Azure Synapse Analytics (formerly SQL Data Warehouse), resource classes are used to manage memory allocation and concurrency for different workloads. The scenario involves ensuring automated data loads complete successfully despite concurrent ad hoc queries consuming resources.

Why Option C is Correct

• Assigning a larger resource class to the automated data load queries ensures they receive higher priority for memory allocation and compute resources
• Larger resource classes (like largerc or xlargerc) provide more memory per query, which is essential for data loading operations that often involve memory-intensive operations like data transformation, sorting, and loading
• This approach allows the automated data loads to complete quickly and successfully even when competing with ad hoc queries for resources
• Resource classes directly control the amount of memory allocated to each query, making this the most targeted solution for memory availability concerns

Why Other Options Are Less Suitable

Option A (Hash distribute large fact tables):

• While hash distribution can improve query performance, it doesn't directly address memory contention between concurrent workloads
• This is a table design consideration rather than a workload management solution

Option B (Assign smaller resource class):

• This would worsen the problem by giving data loads even less memory, making them more likely to fail or run slowly when competing with ad hoc queries

Option D (Create sampled statistics):

• Statistics help with query optimization but don't manage resource allocation between competing workloads
• This addresses query performance optimization rather than memory contention resolution

Best Practice Consideration

In Azure Synapse Analytics, using workload management through resource classes is the recommended approach for prioritizing critical operations like automated data loads over less critical ad hoc queries. This ensures business-critical ETL processes complete reliably while maintaining system availability for user queries.