Architecture patterns are where individual decisions become complete systems.
While pillars explain how to make specific decisions and topics organize what decisions you’ll face, patterns show how everything fits together in production-ready architectures.
These aren’t theoretical patterns. They’re the recurring enterprise architectures that AWS Solutions Architects design repeatedly—and that SAP-C02 tests repeatedly.
The Enterprise Data Lake pattern focuses on breaking down data silos to provide a centralized, secure, and searchable repository for all types of data at scale. For the SAP-C02 exam, you must demonstrate proficiency in designing “decoupled” analytics pipelines that separate storage, processing, and consumption.
🏗️ Core Architectural Patterns #
SAP-C02 expects architects to choose the right tool for each stage of the data lifecycle: Ingestion → Storage → Cataloging → Analytics.
1. The S3-Centric Data Lake #
- Amazon S3: The foundation of the data lake. Key exam focus: Partitioning strategies, Lifecycle policies (S3 Intelligent-Tiering), and Object tagging for security.
- AWS Lake Formation: Simplifies the setup of a secure data lake by providing fine-grained access control (column-level security) across multiple accounts.
2. Processing & Cataloging #
- AWS Glue: Managed ETL (Extract, Transform, Load) and the Glue Data Catalog, which provides a unified metadata repository.
- Amazon EMR: Used for massive parallel processing (Hadoop/Spark) when fine-grained cluster control is required.
- Amazon Athena: Serverless interactive querying using standard SQL directly on S3 data.
3. Purpose-Built Analytics #
- Amazon Redshift (RA3 Instances): For high-performance data warehousing. Focus on Redshift Spectrum for querying S3 data without loading it.
- Amazon Kinesis: For real-time streaming ingestion (Data Streams vs. Firehose).
📚 Key Decision Pillars (Deep Dives) #
These pillars provide the tactical logic for data-heavy SAP-C02 scenarios:
- Enterprise S3 Security & Lifecycle: Optimizing cost with Glacier and securing data with KMS.
- High-Performance Storage: EFS vs FSx: Choosing storage for big data compute nodes.
- Multi-Account Cost Analysis via CUR: Using Athena to analyze organization-wide spending.
- Scaling SQL: RDS Multi-AZ vs Replicas: Understanding data consistency in analytics reporting.