RESEARCH PAPER: Storage-Optimized Machine Learning

By Steve McDowell - March 1, 2018
Organizations run at the speed of their data. While Artificial Intelligence (AI) and Machine Learning (ML) have a continuing history of solving traditional problems in pattern recognition, AI and ML techniques are rapidly finding their place in business analytics, where the patterns being determined might be less obvious. The efficiency of these learning systems can define an organization’s competitive advantage. You can download the paper here.

Table of Contents:

  • Summary
  • The Business Of Machine Learning
  • Machine Learning In Today's World
  • Challenges Of Machine Learning And Storage
  • Architecting For Efficiency
  • Conclusion
  • Figure 1: Modern Learning Pipeline
  • Figure 2: Typical Machine Learning Pipeline
  • Figure 3: GPU Processing VS Traditional Processing
  • Figure 4: New Storage Hierarchy

Companies Cited:

  • Amazon
  • Facebook
  • NVIDIA
  • Pure Storage
+ posts