Paper Info
Title
VLBI-resolution radio-map algorithms: Performance analysis of different levels of data-sharing on multi-socket, multi-core architectures.
Abstract
A broad area in astronomy focuses on simulating extragalactic objects based on Very Long Baseline Interferometry (VLBI) radio-maps. Several algorithms in this scope simulate what would be the observed radio-maps if emitted from a predefined extragalactic object. This work analyzes the performance and scaling of this kind of algorithms on multi-socket, multi-core architectures. In particular, we evaluate a sharing approach, a privatizing approach and a hybrid approach on systems with complex memory hierarchy that includes shared Last Level Cache (LLC). In addition, we investigate which manual processes can be systematized and then automated in future works. The experiments show that the data-privatizing model scales efficiently on medium scale multi-socket, multi-core systems (up to 48 cores) while regardless of algorithmic and scheduling optimizations, the sharing approach is unable to reach acceptable scalability on more than one socket. However, the hybrid model with a specific level of data-sharing provides the best scalability over all used multi-socket, multi-core systems.
Year
DOI
Venue
2012
10.1016/j.cpc.2012.04.016
Computer Physics Communications
Keywords
Field
DocType
VLBI-resolution radio-map algorithms,Parallel approaches,Multi-socket multi-core systems,UMA,NUMA,NUCA
Mathematical optimization,Memory hierarchy,Scheduling (computing),Cache,Computer science,Data sharing,Parallel computing,Very-long-baseline interferometry,Computer engineering,Scaling,Multi-core processor,Scalability
Journal
Volume
Issue
ISSN
183
9
0010-4655
Citations 
PageRank 
References 
0
0.34
3
Authors
5
Name
Order
Citations
PageRank
Siham Tabik116718.92
Luis F. Romero2328.89
Petar Mimica300.68
Oscar G. Plata417330.59
Emilio L. Zapata5811100.36