Comparing CPUs and GPUs for Deep Learning and Artificial Intelligence

Comparing CPUs and GPUs for Deep Learning and Artificial Intelligence
Comparing CPUs and GPUs for Deep Learning and Artificial Intelligence
White Paper
March 30, 2021

Comparing CPUs and GPUs for Deep Learning and Artificial Intelligence

At its Data-Centric Innovation Day in April 2019, Intel introduced its new 2nd Generation Xeon® Scalable Processors. Equipped with features for machine learning and artificial intelligence (AI), such as Intel Deep Learning Boost (Intel DL Boost) and Vector Neural Network Instruction (VNNI), these CPUs strengthened Intel’s play in an increasingly emerging technology space where NVIDIA GPUs have held the dominant position.

In this white paper, we examine Intel’s 2nd Generation Xeon Scalable Processors with NVIDIA’s V100 and T4 GPUs, looking at comparisons using three different benchmarks and extrapolating to understand both technologies' potential in defense and aerospace AI-based applications.

Login and download the white paper to explore

  • Intel CPUs’ and NVIDIA GPUs’ capabilities for deep learning
  • Inference performance comparisons using ResNet-50, BERT, and NCF benchmarks
  • Whether a CPU or GPU is better for aerospace and defense AI applications