ACA-CloudNative 無料問題集「Alibaba Cloud ACA Cloud Native Certification」

（A）AI/ML job lifecycle management

（B）Accelerated computation with distributed data cache

（C）Optimized scheduler for AI/ML jobs

（D）Cost management

（E）GPU monitoring and problem detection

（F）Auto Scaling

（G）Unified heterogeneous resources management

（A）Sidecar

（B）Deployment

（C）StatefulSet

（D）DeamonSet

（A）Microservice amount * Instance type (e.g. 2C/4G)

（B）Microservice amount * Average instance amount * Instance type (e.g. 2C/4G) * Redundancy + Autoscale nodes instance type

（C）Microservice amount * Average instance amount * Instance type (e.g. 2C/4G) * Redundancy

（D）Microservice amount * Average instance amount * Instance type (e.g. 2C/4G)

（A）You should choose higher instance specifications, because you can run more tasks on a single worker node this way, improving utilization.

（B）Attach data disk to worker nodes, because Docker image, system logs, and temp data will consume space. Without a data disk, systems can quickly run out of disk space.

（C）Ensure redundant worker node capacity for each running pod.

（D）You should choose the lowest possible instance specification for your workers, to reduce costs.

（E）If worker node capacity is more than 1000 cores, consider using EBM (ECS Bare Metal).

（A）False

（B）True

（A）Helm Charts can achieve "immutable infrastructure"

（B）Help Charts make it easy to achieve CD (continuous deployment) for applications

（C）Helm Charts make it easy to install Kubernetes objects, such as ConfigMap, Services, Pods, etc.

（D）Help Charts are declarative but not procedural

（A）False

（B）True

（A）False

（B）True

（A）Optimized scheduler for AI/ML jobs

（B）Cost management

（C）Al ML job lifecycle management

（D）GPU monitoring and problem detection

（E）Auto Scaling

（F）Accelerate computation with distributed data cache

（G）Unified heterogeneous resources management

（A）That will saving cost

（B）That will improve performance under ML job

（C）That will accelerate speed communication between each node