Skip to Main content Skip to Navigation
Journal articles

Adaptive Sampling for Performance Characterization of Application Kernels

Abstract : Characterizing performance is essential to optimize programs and architectures. The open source Adaptive Sampling Kit (ASK) measures the performance trade-off in large design spaces. Exhaustively sampling all sets of parameters is computationally intractable. Therefore, ASK concentrates exploration in the most irregular regions of the design space through multiple adaptive sampling strategies. The paper presents the ASK architecture and a set of adaptive sampling strategies, including a new approach called Hierarchical Variance Sampling. ASK's usage is demonstrated on three performance characterization problems: memory stride accesses, Jacobian stencil code, and an industrial seismic application using 3D stencils. ASK builds accurate models of performance with a small number of measures. It considerably reduces the cost of performance exploration. For instance, the Jacobian stencil code design space, which has more than 31 × 10^8 combinations of parameters, is accurately predicted using only 1500 combinations.
Complete list of metadata
Contributor : Pablo de Oliveira Castro Connect in order to contact the contributor
Submitted on : Wednesday, February 26, 2014 - 2:45:19 PM
Last modification on : Wednesday, October 20, 2021 - 12:24:14 AM

Links full text




Pablo de Oliveira Castro, Eric Petit, Asma Farjallah, William Jalby. Adaptive Sampling for Performance Characterization of Application Kernels. Concurrency and Computation: Practice and Experience, Wiley, 2013, 25 (17), pp.2345-2362. ⟨10.1002/cpe.3097⟩. ⟨hal-00952288⟩



Les métriques sont temporairement indisponibles