CAIC Cloud Infrastructure 2 — Questions and Answers
Question 1: Which cloud deployment model gives an organization the most control over its AI infrastructure while still leveraging cloud scalability?
- Public cloud
- Private cloud (Correct answer)
- Hybrid cloud
- Community cloud
Correct answer: Private cloud
Private cloud gives organizations full control over hardware, networking, and security while still using cloud-style orchestration and scalability.
Question 2: An AI workload requires low-latency access to a large dataset stored in object storage. Which architectural pattern best addresses this?
- Cache the dataset in a distributed in-memory store close to compute nodes (Correct answer)
- Move all data to a relational database
- Replicate data across all global regions
- Use a serverless function to fetch data on demand
Correct answer: Cache the dataset in a distributed in-memory store close to compute nodes
Placing a distributed in-memory cache (e.g., Redis or Memcached) adjacent to compute nodes minimizes latency for frequent dataset access.
Question 3: What is the primary advantage of using spot (preemptible) instances for AI training jobs?
- Guaranteed SLA uptime
- Significantly lower cost compared to on-demand instances (Correct answer)
- Faster GPU clock speeds
- Automatic hyperparameter tuning
Correct answer: Significantly lower cost compared to on-demand instances
Spot instances can cost 60–90% less than on-demand instances, making them attractive for fault-tolerant, checkpoint-enabled training jobs.
Question 4: A team is deploying an AI inference service that must handle unpredictable traffic spikes. Which cloud feature is MOST appropriate?
- Reserved instances
- Auto-scaling groups with load balancers (Correct answer)
- Single large VM with fixed resources
- Batch processing queues
Correct answer: Auto-scaling groups with load balancers
Auto-scaling groups dynamically add or remove instances based on demand, ensuring availability during spikes while controlling costs during low traffic.
Question 5: Which storage class is most cost-effective for storing AI training data that is accessed once per month for retraining?
- Standard (hot) storage
- Infrequent access (warm) storage (Correct answer)
- Archive (cold) storage
- Local SSD storage
Correct answer: Infrequent access (warm) storage
Infrequent access storage tiers (e.g., AWS S3-IA, Azure Cool Blob) balance retrieval speed with lower storage costs for data accessed monthly.
Question 6: In a multi-cloud AI strategy, what is the biggest operational challenge?
- Higher total compute capacity
- Vendor lock-in with proprietary services
- Data gravity and consistent governance across providers (Correct answer)
- Unlimited free egress bandwidth
Correct answer: Data gravity and consistent governance across providers
Data gravity—the cost and complexity of moving large datasets—combined with inconsistent IAM and compliance policies is the dominant multi-cloud challenge.
Question 7: Which Kubernetes resource object is used to ensure a specific number of AI inference pod replicas are always running?
- ConfigMap
- ReplicaSet (Correct answer)
- PersistentVolumeClaim
- ServiceAccount
Correct answer: ReplicaSet
A ReplicaSet (or Deployment wrapping one) continuously monitors and maintains the desired number of pod replicas for high availability.
CAIC Cloud Infrastructure 2
Which cloud deployment model gives an organization the most control over its AI infrastructure while still leveraging cloud scalability?