Certified Data Lakehouse Engineer
Audience
The exam tests the technical skills and architectural knowledge required by Data Lakehouse Engineers to successfully implement, manage, and optimize open table formats and distributed scalable storage tiers. A candidate who passes this exam demonstrates a thorough understanding of ACID transactions, file-level metadata pruning, partition evolution, and cluster-wide distributed parallel operations.
Exam Details
- Number of questions: 60
- Duration: 90 minutes
- Pass Score: 60%
- Delivery: Online, proctored.
- Allowed resources: None. You may not use reference materials, white papers, user guides, or any other resources during your exam.
Skills & Knowledge Measured
This exam measures the domain proficiencies and knowledge topics listed in Table 1 below, along with the specific question weight distributions across the exam blueprint.
| Topic | WEIGHT (% of exam) |
|---|---|
1. Transactional Open Table Architecture (Apache Iceberg) |
40% |
2. High-Scale Enterprise Storage Layouts (Apache Ozone) |
35% |
3. Distributed Administrative Utilities & Data Operations |
25% |
Table 1: Exam topics and weighting.
Detailed Syllabus Objectives
Domain 1: Transactional Open Table Architecture (Apache Iceberg) — 40%
Describe the functional architecture of Iceberg’s metadata layers: * Metadata JSON files (centralized history pointers).
Manifest Lists (snapshot composition trackers).
Manifest Files (granular file paths and column-level metrics).
Identify where specific pruning and evaluation statistics are recorded: * Differentiate between row counts, null bounds, and NaN counts tracked inside manifest details vs. raw footers.
Analyze query planning optimization scaling factors: * Compare constant-time O(1) query planning using file-level summaries against legacy linear-time O(N) folder directory scans.
Evaluate row-level mutation and data deletion strategies: * Contrast Copy-on-Write (CoW) hard deletes with lazy Merge-on-Read (MoR) deletion masks.
Explain the mechanics of partition evolution and hidden partitioning: * Describe how Iceberg automatically prunes non-matching files without forcing users to supply explicit, derived partition columns in SQL filters.
Implement data quality assurance patterns and disaster recovery workflows: * Detail the execution phases of the Write-Audit-Publish (WAP) design pattern.
Apply time travel queries for audit compliance and invoke atomic snapshot rollbacks for instantaneous table restoration.
Domain 2: High-Scale Enterprise Storage Layouts (Apache Ozone) — 35%
Describe the components and caching behaviors of the namespace layer: * Explain how the Ozone Manager (OM) utilizes disk-backed RocksDB instances to scale namespace structures while holding only the active metadata working set in physical memory cache.
Identify the functional divisions of the logical storage hierarchy: * Distinguish structural isolation boundaries between Volumes (admin boundaries), Buckets (user storage limits), and Keys (individual dataset objects).
Demonstrate correct storage interface configuration and URI path resolution: * Construct fully qualified data links using the unified Hadoop Compatible FileSystem gateway ().
Identify missing parameters when verifying that a path structure embeds both the target Volume and Bucket namespaces correctly.
Domain 3: Distributed Administrative Utilities & Data Operations — 25%
Analyze cluster-wide parallel data migration strategies: * Evaluate performance constraints when copying datasets between storage environments.
Identify how to parallelize replication workflows using Distributed Copy () to eliminate single-threaded transfer limits.
Apply native command-line parameters to interrogate storage infrastructures: * Determine the correct syntax sequences within the native storage shell framework ().
Distinguish between targeted administrative actions used to query distinct volume or bucket properties (e.g., ).
Suggested Training
Data Lakehouse Engineering Fundamentals
Advanced Architecture: Deep Dive into Apache Iceberg Manifests
Enterprise Object Storage Implementation with Apache Ozone
Helpful Documentation
Iceberg Table Spec Tier Mapping
Ozone Manager RocksDB Architecture Guides
Hadoop-Compatible FileSystem Configuration Protocols