Definition
Livermore Loops are a set of 24 Fortran DO-loops (The Livermore Fortran Kernels, LFK) extracted from operational codes used at the Lawrence Livermore National Laboratory [1, 2]. They have been used since the early 1970s to assess the arithmetic performance of computers and their compilers. They are a mixture of vectorizable and nonvectorizable loops and test rather fully the computational capabilities of the hardware as well as the skill of the software in compiling and vectorization of efficient code. The main value of the benchmark is the range of performance that it demonstrates, and in this respect it complements the limited range of loops tested in the LINPACK benchmark.
Discussion
As a benchmark, the Livermore Loops provide the individual performance of each loop, together with various averages (arithmetic, geometric, harmonic) and the quartiles of the distribution. However, it is difficult to give a clear meaning to these averages, and the value of the benchmark is...
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Bibliography
McMahon FH (1986) The Livermore FORTRAN kernels: a computer test of the numerical performance range. Lawrence Livermore Laboratory technical report LLNL UCRL-53724. Lawrence Livermore National Laboratory, Livermore
McMahon FH (1988) The Livermore Fortran kernels test of the numerical performance range. In: Martin JL (ed) Performance evaluation of supercomputers. Elsevier Science, North-Holland, Amsterdam, pp 143–186
Hockney RW (1991) Performance parameters and benchmarking of supercomputers. Parallel Comput 17:1111–1130
Barley D, Brooks E, Dongarra J, Hayes A, Heath M, Lyon G (1988) Benchmarks to supplant export “FPDR” calculations. Technical report RNR-88-007, National Bureau of Standards, Gaithersburg
Lubeck OM (1988) Supercomputer performance: The theory, practice and results. Los Alamos National Laboratory Report LA-11204-MS, 58 pp
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this entry
Cite this entry
Dongarra, J., Luszczek, P. (2011). Livermore Loops. In: Padua, D. (eds) Encyclopedia of Parallel Computing. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-09766-4_161
Download citation
DOI: https://doi.org/10.1007/978-0-387-09766-4_161
Publisher Name: Springer, Boston, MA
Print ISBN: 978-0-387-09765-7
Online ISBN: 978-0-387-09766-4
eBook Packages: Computer ScienceReference Module Computer Science and Engineering