Efficient Reasoning About Stencil Programs Using Selective Direct Evaluation

04/09/2020
by   Arnab Das, et al.
0

We present FPDetect, a low overhead approach for detecting logical errors and soft errors affecting stencil computations without generating false positives. We develop an offline analysis that tightly estimates the number of floating-point bits preserved across stencil applications. This estimate rigorously bounds the values expected in the data space of the computation. Violations of this bound can be attributed with certainty to errors. FPDetect helps synthesize error detectors customized for user-specified levels of accuracy and coverage. FPDetect also enables overhead reduction techniques based on deploying these detectors coarsely in space and time. Experimental evaluations demonstrate the practicality of our approach.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

04/09/2020

FPDetect: Efficient Reasoning About Stencil Programs Using Selective Direct Evaluation

We present FPDetect, a low overhead approach for detecting logical error...
05/27/2021

Rigorous Roundoff Error Analysis of Probabilistic Floating-Point Computations

We present a detailed study of roundoff errors in probabilistic floating...
05/29/2017

Finding Root Causes of Floating Point Error with Herbgrind

Floating point arithmetic plays a central role in science, engineering, ...
11/06/2020

Low-Cost Floating-Point Processing in ReRAM for Scientific Computing

We propose ReFloat, a principled approach for low-cost floating-point pr...
03/26/2018

Reactive NaN Repair for Applying Approximate Memory to Numerical Applications

Applications in the AI and HPC fields require much memory capacity, and ...
05/31/2020

Improved stochastic rounding

Due to the limited number of bits in floating-point or fixed-point arith...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.