skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: AD-Suite - A Test Suite for Algorithmic Differentiation

Authors:
; ; ;
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Basic Energy Sciences (BES) (SC-22)
OSTI Identifier:
1365844
DOE Contract Number:
AC02-06CH11357
Resource Type:
Conference
Resource Relation:
Conference: 7th International Conference on Algorithmic Differentiation, 09/12/16 - 09/15/16, Oxford, GB
Country of Publication:
United States
Language:
English

Citation Formats

Narayanamurthi, M., Bosse, T., Narayanan, S., and Hovland, P. AD-Suite - A Test Suite for Algorithmic Differentiation. United States: N. p., 2016. Web.
Narayanamurthi, M., Bosse, T., Narayanan, S., & Hovland, P. AD-Suite - A Test Suite for Algorithmic Differentiation. United States.
Narayanamurthi, M., Bosse, T., Narayanan, S., and Hovland, P. 2016. "AD-Suite - A Test Suite for Algorithmic Differentiation". United States. doi:. https://www.osti.gov/servlets/purl/1365844.
@article{osti_1365844,
title = {AD-Suite - A Test Suite for Algorithmic Differentiation},
author = {Narayanamurthi, M. and Bosse, T. and Narayanan, S. and Hovland, P.},
abstractNote = {},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = 2016,
month = 9
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • The compositionally zoned Tinemaha and McMurry Meadows plutons have been studied to determine the compositional variations which occur in sphene, allanite, zircon, apatite, +/- hornblende and +/- augite. Mineral separates (42) were analyzed by INAA from seven granites and a mafic inclusion. The REE, U and Th contents and the mineral/rock ratio of all phases generally increase with increasing whole rock silica content. In all the granite samples (except one) and in the mafic inclusion, sphene is the predominant site for REE, U and Th. In one sample where sphene is interstitial, hornblende is the major REE site. Sphene exhibitsmore » large compositional variations with 2100 to 5800ppm Nd, 25 to 145 ppm Lu, 400 to 1000ppm Th, and 100 to 700ppm Ta. Apatite exhibits large variations in REE (Nd=240 to 1440ppm) and Th (32 to 125ppm). The major compositional variations for zircon are for U (326 to 5610ppm), Th (322 to 6800ppm) and LREE (La=60 to 12000ppm). Hornblende exhibits large variations in REE concentrations (Ce=52 to 214ppm; Yb=1 to 18 ppm). Compositional variations in these minerals can be related to relative variations in the paragenetic sequence and not to either mineral-melt or subsolidus re-equilibration. Indications of the paragenetic sequence of these phases are also preserved in REE zoning (core-rim) of sphene and allanite, determined by both ion and electron microprobes. Therefore, the trace element concentrations of these mineral phases are apparently governed by mineral-melt partitioning during magmatic crystallization and are not subject to significant re-equilibration in the magma or subsolidus.« less
  • We have built a benchmark suite including a number of scientific applications, named test suite, for study of compilation and optimization techniques. Currently, we have Gaussian elimination, FFT, N-body simulation, and other codes in the test suite. We present in this paper the techniques of optimizing program partitioning. Gaussian elimination is used as an example to show how to select the best partitioning pattern with performance evaluation.
  • As scientific codes become more complex and involve larger numbers of developers and algorithms, chances for algorithmic implementation mistakes increase. In this environment, code verification becomes essential to building confidence in the code implementation. This paper will present first results of a new code verification effort within LLNL's B Division. In particular, we will show results of code verification of the LLNL ASC ARES code on the test problems: Su Olson non-equilibrium radiation diffusion, Sod shock tube, Sedov point blast modeled with shock hydrodynamics, and Noh implosion.
  • As parallel systems are commonly being built out of increasingly large multicore chips, application programmers are exploring the use of hybrid programming models combining MPI across nodes and multithreading within a node. Many MPI implementations, however, are just starting to support multithreaded MPI communication, often focussing on correctness first and performance later. As a result, both users and implementers need some measure for evaluating the multithreaded performance of an MPI implementation. In this paper, we propose a number of performance tests that are motivated by typical application scenarios. These tests cover the overhead of providing the MPI{_}THREAD{_}MULTIPLE level of threadmore » safety for user programs, the amount of concurrency in different threads making MPI calls, the ability to overlap communication with computation, and other features. We present performance results with this test suite on several platforms (Linux cluster, Sun and IBM SMPs) and MPI implementations (MPICH2, Open MPI, IBM, and Sun).« less