Performance evaluation of model‐driven partitioning algorithms for data‐parallel kernels on heterogeneous platforms. Issue 1 (31st March 2019)

• Record Type: Journal Article
• Authors: Rico‐Gallego, Juan A.; Díaz‐Martín, Juan C.; Moreno‐Álvarez, Sergio; Calvo‐Jurado, Carmen; García‐Zapata, Juan L.
• Is Part Of: Computational and mathematical methods. Volume 2:Issue 1(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-03-31
• Subjects: communication optimization -- communication performance models -- data‐parallel kernels -- heterogeneous platforms -- partitioning algorithms; Mathematics -- Data processing -- Periodicals; Numerical analysis -- Periodicals; Numerical analysis; Mathematics -- Data processing; Periodicals; 004.0151
• DOI: 10.1002/cmm4.1017
• Languages: English
• ISSNs: 2577-7408

Abstract

Abstract : Data‐ parallel applications running on heterogeneous high‐performance computing platforms require a nonuniform distribution of the workload between available processes. Data partitioning algorithms are formulated as an optimization problem. Departing from the computational performance models of the processes, the goal is to find the partition that minimizes the communication cost. Traditionally, communication volume is the metric used to guide the partitioning. This metric, however, is unable to capture the complexity of current heterogeneous systems, which show uneven communication channels and execute applications with different communication patterns. In this paper, we discuss the role of analytical communication performance models as a metric in partitioning algorithms. First, we describe a method to programmatically predict the communication cost of a data‐parallel kernel based on the τ ‐Lop analytical model. We show that this figure better captures the communication features of applications and platforms. We present results showing that this approach builds partitions that equal or improve the performance of data parallel applications on heterogeneous platforms with respect to previous volume‐based strategies.