ACA-CloudNative ACA Cloud Native Certification Exam Questions and Answers

Scaling out to 500 Pods takes about 30s

Scaling out to 500 Pods takes about 50s

Scaling out to 500 Pods takes about 90s

Heterogeneous computing is supported.

All major machine learning frameworks are supported.

Optimized K8S scheduling is provided for jobs running on the GPU. For example, it can ensure

pods are scheduled on the GPU node, provide shared GPU resources, isolate GPU resources, and do

co-scheduling.

GPU and shared GPU utilization can be monitored via Prometheus on Alibaba Cloud.

The "Arena" tool is developed by the community but integrated with Alibaba Cloud ACK.

Alibaba Cloud ElasticSearch provide self-build Analyser

Alibaba Cloud ElasticSearch cluster can be expanded

Alibaba Cloud ElasticSearch cluster can be downsized

You can ssh your Alibaba Cloud ElasticSearch cluster

Al

DeepLearning

Batching compute

HPC

Supports policy-based global distribution

Global distribution is automated

It provides a high success rate and low latency

Global distribution needs to be triggered manually

True

False

Accelerates remote data reading for AI training jobs, and reduces GPU/CPU waiting time

Supports different storage backends including OSS/HDFS/NAS

Multiple cache layers, including RAM/SSD/HDD

Supports preload once, and reads by many jobs

Supports horizontal scaling

Replicates 3 copies of local data to avoid data loss

Composable workflow, support continuous training pipelines,

Scale out training out to hundreds of nodes

Support heterogeneous resources management, such as GPU, FPGA, RDMA.

All resources are free

Supports continuous training pipelines with composable workflow

Scales out to hundreds of nodes to support AI/ML model training

Supports heterogeneous resource management, including management of GPU, FPGA, and

RDMA-enabled resources

All resources are free

Accelerate remote data reading for Al training jobs, reduce GPU/CPU waiting time

Supports different storage backends including OSS/HDFS/NAS

Multiple cache layers including RAM/SSD/HDD

Supports preload once, read by many jobs

Supports Horizontal scaling

Functionbeat

Metricbeat

Filebeat

Packetbeat

True

False

Standard CPU type

GPU type

Preemptible Instance

Bare Metal Instance

Terraform can configure a Log Project and Log Store in Log Service when creating an ACK

cluster

Terraform can configure kube2ram policy when creating an ACK cluster

Terraform can configure a deployment.yaml file when creating an ACK cluster

Terraform can create a ARMS Helm Chart

True

False

Arena is a CLI tools for machine learning achieve Al task scheduling

Arena is a machine learning framework for Al

Arena developed by alibaba cloud and contribute to kubeflow

Arena developed by alibaba cloud and contribute to kubernetes

sls_ingress_qps

sls_ingress_latency_avg

sls_ingress_latency_99

sls_ingress_pv

True

False

Alibaba Cloud provides 3 levels of security: Infrastructure, Software supply chain, and Runtime

Security

At the Infrastructure layer, Alibaba Cloud Container Service for Kubernetes (ACK) integrates

RAM to achieve authN (Authentication) and RBAC authZ (Authorization)

For Software Supply Chain, Alibaba Cloud provides Image Signing and Image Scanning to

make docker images are secure

Fur Runtime Security, ACK provides network policy based on Flannel and Terway networks

Security Center can defend application runtimes on ACK, and it also can detect attacks within

ACK.

ACK supports Sandboxed containers, which provide an independent kernel to enable enhanced security isolation

True

False