ASK: Adaptive Sampling Kit for Performance Characterization - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Lecture Notes in Computer Science Année : 2012

ASK: Adaptive Sampling Kit for Performance Characterization

Résumé

Characterizing performance is essential to optimize programs and architectures. The open source Adaptive Sampling Kit (ASK) measures the performance trade-offs in large design spaces. Exhaustively sampling all points is computationally intractable. Therefore, ASK concentrates exploration in the most irregular regions of the design space through multiple adaptive sampling methods. The paper presents the ASK architecture and a set of adaptive sampling strategies, including a new approach: Hierarchical Variance Sampling. ASK's usage is demonstrated on two performance characterization problems: memory stride accesses and stencil codes. ASK builds precise models of performance with a small number of measures. It considerably reduces the cost of performance exploration. For instance, the stencil code design space, which has more than 31.10^8 points, is accurately predicted using only 1500 points.

Dates et versions

hal-00952307 , version 1 (26-02-2014)

Identifiants

Citer

Pablo de Oliveira Castro, Eric Petit, Jean Christophe Beyler, William Jalby. ASK: Adaptive Sampling Kit for Performance Characterization. Euro-Par 2012 Parallel Processing, 2012, Greece. pp.89-101, ⟨10.1007/978-3-642-32820-6_11⟩. ⟨hal-00952307⟩

Collections

CNRS UVSQ LI-PARAD
135 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More