Extracting Clean Performance Models from Tainted Programs

12/31/2020
by   Marcin Copik, et al.
0

Performance models are well-known instruments to understand the scaling behavior of parallel applications. They express how performance changes as key execution parameters, such as the number of processes or the size of the input problem, vary. Besides reasoning about program behavior, such models can also be automatically derived from performance data. This is called empirical performance modeling. While this sounds simple at the first glance, this approach faces several serious interrelated challenges, including expensive performance measurements, inaccuracies inflicted by noisy benchmark data, and overall complex experiment design, starting with the selection of the right parameters. The more parameters one considers, the more experiments are needed and the stronger the impact of noise. In this paper, we show how taint analysis, a technique borrowed from the domain of computer security, can substantially improve the modeling process, lowering its cost, improving model quality, and help validate performance models and experimental setups.

READ FULL TEXT

Authors

page 7

page 10

03/04/2016

Performance Localisation

Performance becomes an issue particularly when execution cost hinders th...
06/01/2019

A Technique for Finding Optimal Program Launch Parameters Targeting Manycore Accelerators

In this paper, we present a new technique to dynamically determine the v...
11/05/2019

KLARAPTOR: A Tool for Dynamically Finding Optimal Kernel Launch Parameters Targeting CUDA Programs

In this paper we present KLARAPTOR (Kernel LAunch parameters RAtional Pr...
05/09/2022

The Role of Idle Waves, Desynchronization, and Bottleneck Evasion in the Performance of Parallel Programs

The performance of highly parallel applications on distributed-memory sy...
06/02/2022

Noise2NoiseFlow: Realistic Camera Noise Modeling without Clean Images

Image noise modeling is a long-standing problem with many applications i...
05/11/2020

Validation of Abstract Side-Channel Models for Computer Architectures

Observational models make tractable the analysis of information flow pro...
06/26/2019

Time-evolving psychological processes over repeated decisions

Many psychological experiments have participants repeat a simple task. T...
This week in AI

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