Describe Convergence of AI, High-Performance Computing, and Analytics CERTIFIED-AI-EXPERT Practice Questions (Page 6)

Questions 26–30

1. 26application · medium

   A hospital wants to use patient data to not only identify which patients are at high risk of readmission but also to recommend personalized care plans to reduce that risk. They currently have a data warehouse with historical patient records. What should they add to achieve this?

   • AMore detailed data entry forms to capture additional patient information for the warehouse
   • BA high-performance computing cluster to run complex simulations of patient outcomes
   • CA machine learning model that predicts readmission risk and suggests interventions based on patient profiles
   • DA new dashboard that displays readmission rates by department
   Reveal Solution Discussion Select an answer first
2. 27application · medium

   A logistics company processes GPS data from thousands of vehicles to optimize delivery routes. The data volume is huge, and the current analytics software runs on a single server, causing delays. They need to process the data faster. What is the best way to use HPC to improve their analytics?

   • AUse a distributed computing framework to process the GPS data in parallel across multiple nodes
   • BUse a machine learning model to predict routes without processing all the GPS data
   • CStore the GPS data in a faster storage system to speed up the single server
   • DCompress the GPS data to reduce its size so the single server can process it faster
   Reveal Solution Discussion Select an answer first
4. 28foundation · easy

   How does HPC enable large-scale data processing for analytics?

   • AHPC stores data in a data warehouse
   • BHPC automatically cleans and prepares data for analysis
   • CHPC provides the computational power to process massive datasets quickly
   • DHPC creates data visualizations for analysts
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A telecommunications company analyzes network traffic logs to detect anomalies. The data volume grows daily, and their current analytics server cannot keep up with the processing demand. They need a solution that can scale out to handle the increasing data. What is the best approach?

   • AUpgrade the existing server with more CPUs and RAM
   • BStore the logs in a data lake and query them only when needed
   • CDeploy a distributed analytics framework (e.g., Spark) on a cluster of commodity servers
   • DUse a machine learning model to filter out normal traffic so only anomalies are processed
   Reveal Solution Discussion Select an answer first
6. 30expert · hard

   An energy company wants to optimize wind turbine performance. They have sensors generating terabytes of data daily, need to run computational fluid dynamics (CFD) simulations to test new blade designs, and want to use historical data to predict maintenance needs. They have a limited budget and must choose a single platform. Which platform best supports the unified convergence of AI, HPC, and analytics?

   • AA data lake with a SQL engine and a separate AI service, but no HPC cluster
   • BA cloud-based analytics service that can run simple simulations and train small AI models
   • CA dedicated AI supercomputer that can also run analytics, but not CFD simulations
   • DA converged HPC cluster with integrated analytics tools and AI frameworks, shared across teams
   Reveal Solution Discussion Select an answer first