[AZ-104] Azure Synapse Analytics 실전 모의문제 — 05/23

작성자: azure | 작성일: 2026년 05월 23일 | 조회: 2 | 좋아요: 0

⚙️
ASSOCIATE LEVEL

Microsoft Azure Administrator Associate

AZ-104 | 다중 선택 | 데이터 분석

📝 QUESTION

Contoso Manufacturing operates a global network of factories and relies heavily on data analytics for operational insights. Their current analytics infrastructure in Azure processes vast amounts of sensor data, historical production records from machines, and supply chain information. They use an always-on **Azure HDInsight Spark cluster** for daily batch processing of historical data stored in **Azure Data Lake Storage Gen2 (ADLS Gen2)**, **Azure Stream Analytics** for real-time sensor data processing, and **Azure Data Factory** for data ingestion and orchestration. Management has mandated a 25% reduction in their monthly Azure analytics bill without compromising data freshness requirements (daily reports, real-time dashboards). The company's IT operations team, responsible for managing this infrastructure, needs to identify and implement cost-effective solutions. Which three (3) of the following actions should the IT operations team recommend to optimize costs while maintaining performance and data freshness?

A. Replace the always-on Azure HDInsight Spark cluster with an Azure Synapse Analytics Spark pool configured for auto-pause and optimized for cost.
B. Migrate all real-time sensor data processing from Azure Stream Analytics to an Azure Databricks Structured Streaming cluster configured with auto-scaling and auto-termination.
C. Utilize Azure Synapse Analytics Serverless SQL pool for ad-hoc queries and daily report generation directly on data in ADLS Gen2, reducing the need for dedicated SQL pools or additional ETL steps.
D. Implement Azure Data Factory's tumbling window triggers instead of event-based triggers for all batch processing pipelines to reduce execution frequency.
E. Optimize Azure Data Factory pipeline activities by consolidating multiple smaller activities into fewer, larger ones, and leverage Data Factory's auto-scale features for Self-Hosted Integration Runtimes if applicable to reduce VM costs.
⬇️ 아래에서 정답과 해설을 확인하세요 ⬇️
✅ ANSWER
정답: A, C, E
✅ A. Replace the always-on Azure HDInsight Spark cluster with an Azure Synapse Analytics Spark pool configured for auto-pause and optimized for cost.
❌ B. Migrate all real-time sensor data processing from Azure Stream Analytics to an Azure Databricks Structured Streaming cluster configured with auto-scaling and auto-termination.
✅ C. Utilize Azure Synapse Analytics Serverless SQL pool for ad-hoc queries and daily report generation directly on data in ADLS Gen2, reducing the need for dedicated SQL pools or additional ETL steps.
❌ D. Implement Azure Data Factory's tumbling window triggers instead of event-based triggers for all batch processing pipelines to reduce execution frequency.
✅ E. Optimize Azure Data Factory pipeline activities by consolidating multiple smaller activities into fewer, larger ones, and leverage Data Factory's auto-scale features for Self-Hosted Integration Runtimes if applicable to reduce VM costs.
📖 EXPLANATION

A. Azure Synapse Analytics Spark pool은 자동 일시 중지 및 동적 스케일링을 제공하여 상시 가동되는 HDInsight 클러스터보다 비용 효율적입니다. C. Azure Synapse Serverless SQL pool은 쿼리당 비용을 지불하므로 ADLS Gen2의 데이터에 대한 임시 분석 및 보고에 적합하며 전용 컴퓨팅 비용을 절감합니다. E. Azure Data Factory 활동을 통합하면 청구되는 활동 수가 줄어들고, 자체 호스팅 통합 런타임(Self-Hosted Integration Runtimes)의 자동 스케일링 기능은 기본 VM 비용을 최적화할 수 있습니다.

💡 핵심 포인트: 비용 최적화 문제에서는 각 Azure 서비스의 과금 모델과 효율적인 운영 방식을 이해하는 것이 중요합니다. 특히, '항상 켜져 있는(always-on)' 리소스를 '필요할 때만 사용하는(pay-per-use)' 또는 '자동으로 스케일 아웃/인 되는' 서비스로 대체하는 방안을 고려해야 합니다.

🏷️ 관련 Azure 서비스

Azure Synapse AnalyticsAzure HDInsightAzure Data FactoryAzure Data Lake Storage Gen2Azure Stream Analytics

📚 Azure 자격증 준비를 위한 데일리 모의문제

매일 새로운 문제가 업데이트됩니다 | 더 많은 문제 풀기 →