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

Title: Final Report

Abstract

The purpose of this project was to develop tools and techniques to improve the ability of computational scientists to investigate and correct problems (bugs) in their programs. Specifically, the University of Maryland component of this project focused on the problems associated with the finite number of bits available in a computer to represent numeric values. In large scale scientific computation, numbers are frequently added to and multiplied with each other billions of times. Thus even small errors due to the representation of numbers can accumulate into big errors. However, using too many bits to represent a number results in additional computation, memory, and energy costs. Thus it is critical to find the right size for numbers. This project focused on several aspects of this general problem. First, we developed a tool to look for cancelations, the catastrophic loss of precision in numbers due to the addition of two numbers whose actual values are close to each other, but whose representation in a computer is identical or nearly so. Second, we developed a suite of tools to allow programmers to identify exactly how much precision is required for each operation in their program. This tool allows programmers to both verify thatmore » enough precision is available, but more importantly find cases where extra precision could be eliminated to allow the program to use less memory, computer time, or energy. These tools use advanced binary modification techniques to allow the analysis of actual optimized code. The system, called Craft, has been applied to a number of benchmarks and real applications.« less

Authors:
Publication Date:
Research Org.:
University of Maryland
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
OSTI Identifier:
1148696
Report Number(s):
Final Report
DOE Contract Number:
SC0002351
Resource Type:
Technical Report
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING

Citation Formats

Hollingsworth, Jeff. Final Report. United States: N. p., 2014. Web. doi:10.2172/1148696.
Hollingsworth, Jeff. Final Report. United States. doi:10.2172/1148696.
Hollingsworth, Jeff. Thu . "Final Report". United States. doi:10.2172/1148696. https://www.osti.gov/servlets/purl/1148696.
@article{osti_1148696,
title = {Final Report},
author = {Hollingsworth, Jeff},
abstractNote = {The purpose of this project was to develop tools and techniques to improve the ability of computational scientists to investigate and correct problems (bugs) in their programs. Specifically, the University of Maryland component of this project focused on the problems associated with the finite number of bits available in a computer to represent numeric values. In large scale scientific computation, numbers are frequently added to and multiplied with each other billions of times. Thus even small errors due to the representation of numbers can accumulate into big errors. However, using too many bits to represent a number results in additional computation, memory, and energy costs. Thus it is critical to find the right size for numbers. This project focused on several aspects of this general problem. First, we developed a tool to look for cancelations, the catastrophic loss of precision in numbers due to the addition of two numbers whose actual values are close to each other, but whose representation in a computer is identical or nearly so. Second, we developed a suite of tools to allow programmers to identify exactly how much precision is required for each operation in their program. This tool allows programmers to both verify that enough precision is available, but more importantly find cases where extra precision could be eliminated to allow the program to use less memory, computer time, or energy. These tools use advanced binary modification techniques to allow the analysis of actual optimized code. The system, called Craft, has been applied to a number of benchmarks and real applications.},
doi = {10.2172/1148696},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu Jul 31 00:00:00 EDT 2014},
month = {Thu Jul 31 00:00:00 EDT 2014}
}

Technical Report:

Save / Share:
  • A process design for a CLEAN COKE commercial facility has been developed, based on information obtained from completed bench-scale and process-development studies. The design envisages a plant, based on Illinois No. 6 coal, and with a 21-year operating life. A coal feed rate of 12.43 million tons per year, raw basis (8.9 million tons MAF basis), is comprehended. Carbonization encompasses a preoxidation/drying stage followed by two stages of devolatilization to a maximum temperature of 1400/sup 0/ F. This yields a product char with residual volatile-matter and sulfur contents of 2 and 0.78 percent, respectively. The char is converted to amore » pelletized formcoke in a sequence of operations including disk pelletization, air curing, and final coking. Hydrogenation is carried out noncatalytically at 870/sup 0/F and 2700 psia pressure, with a 1-hour residence time. Products of hydrogenation are separated by a vapor-stripping technique, and the carbonaceous residue is processed through a slagging gasifier to produce additional hydrogen for the process. The products from the plant include, on an annual basis, 3.6 million tons of coke, 2.9 billion pounds of chemical feedstocks, and more than 33 trillion Btu of liquid fuel.« less
  • The Western Study Area (WSA) report integrates the study area history, geology, and hydrology with the results of soil, surface water, ground water, air, biota, and structures investigations to define the nature and extent of contamination of the western part of RMA. The WSA includes all of sections 3, 4, 9, and 33 and portions of sections 28 and 34. The WSA is a relatively undeveloped region that was used for railroad transport, vehicle maintenance, and waste disposal. The chemical sampling included analysis of 1081 samples from 388 borings, three soil gas programs, and monitoring of 369 wells. The dominantmore » contaminant group is volatile halogenated organic compounds. The volume of potentially contaminated soil is estimated to be 178,000 cubic yards.« less
  • This report contains two separate parts: Characterization of the Radon Source in North-Central Florida (final report part 1 -- final project report); and Characterization of the Radon Source in North-Central Florida (technical report). The objectives were to characterize the radon 222 source in a region having a demonstrated elevated indoor radon potential and having geology, lithology, and climate that are different from those in other regions of the U.S. where radon is being studied. Radon availability and transport in this region were described. Approaches for predicting the radon potential of lands in this region were developed.
  • The objectives of this work were to determine if the numbér of fiber cell initials varied genetically and to compare the number of initials with that of mature fibers obtained at harvest time. The method used to count the number of fiber cell initials is direct, simple, quick and done while the plant is growing. In contrast, the currently used commercial process is indirect and needs large amount mature fibers gathered at harvest time. However, all current work on cotton yield is based on fiber numbers obtained by the indirect commercial process. Consequently, it was necessary to compare results obtainedmore » from the two methods using the same plants as the source of material. The results show that the number of fiber initials per ovule differed significantly (P>0.05) for seven cultivars in 1995 and 1996. AIso, a 1997 study shows the number of fiber initials varied by 15% over boll positions and environments, with rankings among cultivars generally consistent across boll positions and sampling times. Finally, although there were differences among cultivars for initial fiber cell number, all cultivars had nearly the same number of mature lint fibers per seed. This last finding is significant. It indicates that the rate of fiber cell initiation varies among cultivars; the lower the rate, the greater the difference between the number of initials and the number of mature fiber cells. If the rate of fiber initiation is relatively high, the number of initials and mature fibers differs by about 11%; if it is low, the difference is as high as 31%. Cotton breeders may be able to use genetic differences for the number of fiber initials and/or the rate of fiber cell initiation in crop improvement programs.« less
  • This volume contains: bidding requirements; terms and conditions; specifications for general requirements and sitework; and subcontract drawings.