Benchmarking Data Analysis and Machine Learning Applications on the Intel KNL Many-Core Processor

TL;DR

This paper benchmarks data analysis and machine learning applications on the Intel KNL many-core processor, showing significant performance improvements over previous technologies and other Intel CPUs, highlighting KNL's suitability for these workloads.

Contribution

It provides a comprehensive performance evaluation of data analysis and machine learning applications on the Intel KNL processor, demonstrating notable speedups and efficiency gains.

Findings

DGEMM performance improved by ~3.5x on KNL

Applications achieved ~60% of theoretical peak performance

Caffe machine learning application showed 2.7x speedup on KNL

Abstract

Knights Landing (KNL) is the code name for the second-generation Intel Xeon Phi product family. KNL has generated significant interest in the data analysis and machine learning communities because its new many-core architecture targets both of these workloads. The KNL many-core vector processor design enables it to exploit much higher levels of parallelism. At the Lincoln Laboratory Supercomputing Center (LLSC), the majority of users are running data analysis applications such as MATLAB and Octave. More recently, machine learning applications, such as the UC Berkeley Caffe deep learning framework, have become increasingly important to LLSC users. Thus, the performance of these applications on KNL systems is of high interest to LLSC users and the broader data analysis and machine learning communities. Our data analysis benchmarks of these application on the Intel KNL processor indicate that single-core double-precision generalized matrix multiply (DGEMM) performance on KNL systems has improved by ~3.5x compared to prior Intel Xeon technologies. Our data analysis applications also achieved ~60% of the theoretical peak performance. Also a performance comparison of a machine learning application, Caffe, between the two different Intel CPUs, Xeon E5 v3 and Xeon Phi 7210, demonstrated a 2.7x improvement on a KNL node.