Run-to-Run Reproducibility of Floating-Point Calculations for Applications on Intel® Xeon Phi™ Coprocessors (and Intel® Xeon® Processors)

The Issue

If I rerun the identical program on the identical input data on an identical processor, will I get an identical result?

Authored by Martyn Corden (Intel) Last updated on 06/07/2017 - 10:24

Getting reproducible results with Intel® MKL

This article discusses the underlying reasons for inconsistent results, the mechanisms that cause the variations in Intel MKL, and some ways to improve chances for getting bit to bit correspondence in certain cases.
Authored by TODD R. (Intel) Last updated on 08/22/2017 - 00:27

Consistency of Floating-Point Results using the Intel® Compiler

Why doesn’t my application always give the same answer? Tradeoffs between floating-point accuracy, reproducibility and performance are discussed with examples;recommendations are made for improving reproducibility while limiting the impact on performance. Updated for Intel® Parallel Studio XE 2018, Composer Edition, containing the Intel® Compiler version 18.
Authored by Martyn Corden (Intel) Last updated on 09/11/2017 - 17:53

Introduction to Conditional Numerical Reproducibility (CNR)

Starting with 11.0 release,  Intel® MKL introduces a feature called Conditional Numerical Reproducibility (CNR) which provides functions for obtaining reproducible floating-point results when calli

Authored by TODD R. (Intel) Last updated on 10/26/2017 - 23:06
For more complete information about compiler optimizations, see our Optimization Notice.