ACA CCE study plan

Define the core cloud service models (IaaS, PaaS, SaaS) and map each Alibaba Cloud product family to the model it represents

Explain regional and zone architecture, including how to choose a region for latency, compliance, and disaster recovery

Compare elasticity, scalability, and high availability, and describe the mechanism behind each

Sketch a reference architecture for a simple two-tier web application using only concepts from Domain 1

Describe virtualization, containers, and serverless as deployment models and state one trade-off of each

Explain VPC concepts: subnets, route tables, security groups, and NAT, and trace a packet path from internet to private instance

Identify the shared responsibility boundary between customer and provider for compute, storage, and network

Calculate a simple availability estimate from component redundancy

Configure an ECS instance end to end: instance type selection, image, disk type, security group, and public IP assignment

Describe the purpose and limits of SLB, Auto Scaling, and Object Storage Service in a resilient web tier

Differentiate block storage, object storage, and file storage by access pattern and durability

Explain how a VPC, SLB, and Auto Scaling group combine to serve traffic across zones

Configure a relational database instance with a read-only replica and explain the failover behavior

Describe how ApsaraDB for Redis is used for session caching and state one consistency trade-off

Explain DNS resolution and CDN caching behavior for a static asset

Compare messaging and queue services by delivery guarantee and use case

Differentiate the model service categories (foundation models, embeddings, speech, vision) and state a use case for each

Describe the request lifecycle of a model inference call, including token limits, latency, and cost drivers

Explain how retrieval-augmented generation combines a vector store with a foundation model

Identify where prompt design, temperature, and context length affect output quality

Explain the difference between pre-trained model consumption and model fine-tuning, including data requirements for each

Describe how to evaluate a model output against a rubric and when to reject a response

Identify data privacy obligations when sending customer content to a model service

List the operational controls for rate limiting and fallback when a model endpoint degrades

Configure RAM users, groups, roles, and policies to enforce least privilege for a stated scenario

Explain the role of security groups, network ACLs, and WAF in layered defense

Interpret a billing statement: identify chargeable resources, billing cycle, and one cost-reduction lever

Describe the support plan tiers and state which escalation path applies to a production outage

Re-derive the shared-responsibility boundary and VPC packet path from memory without notes

Reconstruct the three-tier architecture and name the product behind each tier

Explain the RAG request lifecycle and its cost drivers in under three minutes

Resolve ten missed practice questions by writing the correct reasoning for each