Performance evaluation of model‐driven partitioning algorithms for data‐parallel kernels on heterogeneous platforms

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...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computational and mathematical methods Jg. 2; H. 1
Hauptverfasser: Rico‐Gallego, Juan A., Díaz‐Martín, Juan C., Moreno‐Álvarez, Sergio, Calvo‐Jurado, Carmen, García‐Zapata, Juan L.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Hoboken John Wiley & Sons, Inc 01.01.2020
Schlagworte:
Algorithms
Communication
communication optimization
communication performance models
Cost analysis
data‐parallel kernels
heterogeneous platforms
Kernels
Mathematical models
Optimization
Partitioning
partitioning algorithms
Performance enhancement
Performance evaluation
Workload
ISSN:2577-7408, 2577-7408
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung: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.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2577-7408
2577-7408
DOI:10.1002/cmm4.1017