Describe Convergence of AI, High-Performance Computing, and Analytics CERTIFIED-AI-EXPERT Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A financial institution needs to run risk analytics on petabytes of transaction data. The current analytics platform takes days to process the data, and they need results in hours. They have access to a high-performance computing cluster. What is the most effective way to leverage HPC for this analytics workload?

   • AMove the data to a data lake and use a SQL query engine to run the analytics
   • BUse a machine learning model to predict the results of the analytics without processing all the data
   • CParallelize the analytics job across the HPC cluster using distributed computing frameworks like MPI or Spark
   • DIncrease the memory of the single analytics server to hold more data in RAM
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   What does the unified convergence of AI, HPC, and analytics enable?

   • AEnd-to-end data-driven problem solving, from data processing to insight generation
   • BReplacing all human decision-making with automated systems
   • CCreating a single software application for all tasks
   • DEliminating the need for data storage
   Reveal Solution Discussion Select an answer first
4. 13expert · hard

   A smart city project wants to manage traffic flow. They have sensors collecting real-time data, need to run simulations of traffic patterns for planning, and want to use AI to predict congestion and suggest alternative routes. They have a limited IT staff and need a solution that is easy to manage. Which approach best balances the convergence of AI, HPC, and analytics with manageability?

   • AUse a managed cloud platform that offers HPC clusters, analytics services, and AI tools as integrated services
   • BUse only AI to predict traffic and skip simulations and analytics to keep it simple
   • CBuild an on-premises HPC cluster and hire more IT staff to manage it, along with separate analytics and AI tools
   • DUse an analytics platform that can also run small simulations and train simple AI models
   Reveal Solution Discussion Select an answer first
5. 14foundation · easy

   What is the relationship between AI and HPC in a converged environment?

   • AAI replaces HPC in all scientific computing tasks
   • BHPC supplies the computational infrastructure that enables AI to process large-scale problems
   • CHPC is a subset of AI that focuses on hardware
   • DAI is used to cool HPC data centers
   Reveal Solution Discussion Select an answer first
6. 15expert · hard

   A pharmaceutical company uses AI to screen potential drug compounds. The screening model is computationally intensive and currently runs on a small GPU cluster. They have access to an HPC cluster used for molecular dynamics simulations. They want to speed up the AI screening without disrupting the simulation workload. What is the best strategy?

   • AMove all simulations to the cloud to free up the HPC cluster exclusively for AI screening
   • BIntegrate the AI screening into the HPC cluster's job scheduler, running it on idle nodes during off-peak hours
   • CRun the AI screening on the small GPU cluster but increase the batch size to improve throughput
   • DUse the HPC cluster only for simulations and purchase a dedicated AI supercomputer for screening
   Reveal Solution Discussion Select an answer first