How to Evaluate Different Storage Options for Artificial Intelligence Workloads

How to Evaluate Different Storage Options for Artificial Intelligence Workloads

This IDC Perspective discusses the challenge of choosing the right storage infrastructure for artificial intelligence workloads, pointing out considerations and features that can help enterprises select the right storage infrastructure for their artificial intelligence workloads. Artificial intelligence workloads are becoming more popular in enterprises undergoing digital transformation. Although enterprises initially tried to run them on traditional scale-out file system platforms, it is becoming clear that many of them have workload requirements that present real challenges for these legacy architectures."Choosing the right storage infrastructure for artificial intelligence workloads requires an excellent understanding of an enterprise's data pipeline and business requirements," said Eric Burgener, research vice president, Infrastructure Systems, Platforms, and Technologies Group at IDC. "IDC recommends that customers closely investigate five areas — performance, availability and resiliency, flexibility, ease of use, and infrastructure efficiency and cost — before making any 'storage for AI' purchase decisions."

Please Note: Extended description available upon request.


Executive Snapshot
Situation Overview
The Rise of Artificial Intelligence–Driven Workloads
Storage for AI Is Different than Traditional Scale-Out NAS
Data-Driven Next-Generation Applications
Understanding the I/O Profiles in Your Artificial Intelligence Data Pipeline
Performance
Availability and Resiliency
Flexibility
Ease of Purchase, Deployment, Management, and Scaling
Infrastructure Efficiency and Cost
Advice for the Technology Buyer
Learn More
Related Research
Synopsis

Download our eBook: How to Succeed Using Market Research

Learn how to effectively navigate the market research process to help guide your organization on the journey to success.

Download eBook