Examkin

Question

Difficulty: HardData Integration and Analytical Storage

An enterprise is designing a modern data platform on Azure to process and analyze telemetry data. The design must meet the following requirements:
- Data analysts must perform ad-hoc, exploratory queries directly against raw CSV and JSON files stored in the data lake, paying only for the queries run.
- The data storage solution must remain resilient and available even in the event of a primary datacenter outage within the region.
- Shared Access Signatures (SAS) provided to external partners must be easily revocable at any time without rotating the storage account access keys.

Which of the following architectural recommendations should you include in the design? (Select TWO).

  1. Provision Azure Synapse Analytics serverless SQL pools to query the raw CSV and JSON files directly in the data lake.Answer
  2. B
    Provision an Azure Synapse Analytics dedicated SQL pool to store and run exploratory queries on the raw files.
  3. Configure the Azure Data Lake Storage Gen2 account to use Geo-Redundant Storage (GRS) and associate all Shared Access Signatures with stored access policies.Answer
  4. D
    Configure the Azure Data Lake Storage Gen2 account to use Locally Redundant Storage (LRS) and generate ad-hoc Shared Access Signatures with long lifetimes.

Answer

Use Azure Synapse Analytics serverless SQL pools for ad-hoc queries, and configure Azure Data Lake Storage Gen2 with Geo-Redundant Storage (GRS) and stored access policies for SAS delegation.
The correct recommendations are to use serverless SQL pools for querying the data lake directly on an ad-hoc basis, and to use Geo-Redundant Storage (GRS) with stored access policies. Serverless SQL pools charge based on data processed, avoiding idle compute costs. GRS provides regional disaster resilience, and stored access policies allow SAS tokens to be revoked easily without changing the root keys.

Step-by-Step Solution

1
Evaluate the ad-hoc query requirements against Azure Synapse options.
Identify that serverless SQL pools are ideal for ad-hoc queries on raw files because they run on-demand and charge only per query, whereas dedicated SQL pools incur persistent costs.
To satisfy the requirement of minimal overhead and pay-per-query model for ad-hoc exploration.
2
Evaluate redundancy options for regional disaster resilience.
Determine that Geo-Redundant Storage (GRS) replicates data across regions to protect against regional outages, whereas Locally Redundant Storage (LRS) only replicates within a single datacenter.
To satisfy the requirement of surviving a primary datacenter outage.
3
Evaluate security requirements for SAS token revocation.
Identify that associating SAS tokens with a stored access policy allows instant revocation by deleting or modifying the policy, avoiding the need to rotate primary/secondary keys.
To satisfy the requirement of revoking access keys without rotating the storage account access keys.

Key Concept

Designing data integration and analytical storage requires matching query models (serverless vs. dedicated) to usage patterns, and configuring storage redundancy and access control policies to meet resilience and security requirements.
Rate this question

Topics

Data Integration and Analytical Storage

More questions from this topic