Curriculum
ACP-004 — 5-Module Programme
- MaxCompute architecture — projects, tables (internal/external/virtual views), partitioning, schema evolution, and lifecycle management
- MaxCompute SQL — DDL/DML, complex types (ARRAY/MAP/STRUCT), WINDOW functions, LATERAL VIEW, and query optimisation
- MaxCompute MapReduce and Graph computing — job submission, distributed algorithms, and performance tuning
- MaxCompute security — LabelSecurity, Column-level ACL, project-level policy, and compliance with data classification
- MaxCompute Tunnel SDK architecture — upload sessions, block management, checkpointing, and concurrent upload strategies
- Implementing high-throughput bulk data uploads using Tunnel SDK — Java/Python clients, retry strategies, and error handling
- DataHub streaming ingestion — connectors, schema registry, and integration with MaxCompute and E-MapReduce
- DataSync service — synchronisation tasks, data source connectors (RDS, OSS, MongoDB), and incremental load strategies
- DataWorks Studio — nodes, workflows, task dependencies, scheduling, baseline management, and SLA alerting
- Data Integration — Reader/Writer plugins, real-time sync, offline sync, and network environment configuration for private data sources
- Data Quality — DQC rules, blocking rules, monitoring templates, and alert configuration for data pipeline governance
- DataMap — data asset catalogue, data lineage, sensitivity marking, and DataService Studio API publishing
- EMR cluster lifecycle — Hadoop, Spark, Hive, Presto, HBase, Zookeeper, Kafka cluster creation and management on Alibaba Cloud
- EMR data processing — Hive partitioned tables, Spark DataFrames, HBase rowkey design, and Presto federation queries
- EMR with OSS — EMRFS, cold/hot data tiering, and EMR Serverless Spark for elastic job execution
- EMR security — Kerberos authentication, Ranger fine-grained access control, and audit logging for regulated environments
- Alibaba Cloud PAI — Machine Learning Platform overview, feature engineering, model training on GPU, and EAS deployment
- Flink on Alibaba Cloud (Realtime Compute) — streaming jobs, event-time processing, state backends, checkpointing, and SQL mode
- Quick BI — data source connection, dataset modelling, dashboard design, cross-filter interactions, and portal publishing
- End-to-end architecture: Real-time fraud detection using Flink → MaxCompute → PAI → Quick BI dashboard
ACP Big Data Professional
Course & Exam Details
5 Modules
Full ACP-004 domains
120 Minutes
Extended exam duration
60 Questions
MCQ format
65 / 100
Passing score (Professional)
USD 120.00
Exam registration fee
Tech Support
Call / WhatsApp
Mon–Fri
9 AM – 6 PM