181 Matching Results

Search Results

Advanced search parameters have been applied.

Using gpfs 2.2 to enable a cross platform accessibility of singlestorage

Description: With IBM's aid I have conducted a cross compatibility test of GPFS 2.2 between an IBM F50 Power2 running AIX 5.2 ML/3 and 8 Dual Pentium 4/2.2 GHz running Redhat 9.0. The objective was to demonstrate a single shared instance of the file system and storage between the disparate operating systems and hardware systems. The cross compatibility test was successful. The chronology of events that led to this successful test are documented below.
Date: December 1, 1994
Creator: Baird, Will
Partner: UNT Libraries Government Documents Department

An experimental evaluation of the parallel I/O systems of the IBM SP and Intel Paragon using a production application

Description: This paper presents the results of an experimental evaluation of the parallel I/O systems of the IBM SP and Intel Paragon. For the evaluation, we used a full, three-dimensional application code that is in production use for studying the nonlinear evolution of Jeans instability in self-gravitating gaseous clouds. The application performs I/O by using library routines that we developed and optimized separately for parallel I/O on the SP and Paragon. The I/O routines perform two-phase I/O and use the PIOFS file system on the SP and PFS on the Paragon. We studied the I/O performance for two different sizes of the application. We found that for the small case, I/O was faster on the SP, whereas for the large case, I/O took almost the same time on both systems. Communication required for I/O was faster on the Paragon in both cases. The highest read bandwidth obtained was 48 Mbytes/sec. and the highest write bandwidth obtained was 31.6 Mbytes/sec., both on the SP.
Date: September 1, 1996
Creator: Thakur, R.; Gropp, W. & Lusk, E.
Partner: UNT Libraries Government Documents Department

Early Experiences with the IBM SP1 and the High-Performance Switch

Description: The IBM SP1 is IBM`s newest parallel distributed-memory computer. As part of a joint project with IBM, Argonne took delivery of an early system in order to evaluate the software environment and to begin porting programming packages and applications to this machine. This report discusses the results of those efforts once the high-performance switch was installed. An earlier report (ANL/MCS-TM-177) emphasized software usability and the initial ports to the SP1. This report contains performance results and discusses some applications and tools not covered in TM 177.
Date: November 1993
Creator: Gropp, William
Partner: UNT Libraries Government Documents Department

Performance of large-scale scientific applications on the IBM ASCI Blue-Pacific system

Description: The IBM ASCI Blue-Pacific System is a scalable, distributed/shared memory architecture designed to reach multi-teraflop performance. The IBM SP pieces together a large number of nodes, each having a modest number of processors. The system is designed to accommodate a mixed programming model as well as a pure message-passing paradigm. We examine a number of applications on this architecture and evaluate their performance and scalability.
Date: December 10, 1998
Creator: Mirin, A
Partner: UNT Libraries Government Documents Department

Users guide for the ANL IBM SP1

Description: This guide presents the features of the IBM SP1 installed in the Mathematics and Computer Science Division at Argonne National Laboratory. The guide describes the available hardware and software, access policies, and hints for using the system productively.
Date: October 1, 1994
Creator: Gropp, W.; Lusk, E. & Pieper, S. C.
Partner: UNT Libraries Government Documents Department

An efficient parallel algorithm for mesh smoothing

Description: Automatic mesh generation and adaptive refinement methods have proven to be very successful tools for the efficient solution of complex finite element applications. A problem with these methods is that they can produce poorly shaped elements; such elements are undesirable because they introduce numerical difficulties in the solution process. However, the shape of the elements can be improved through the determination of new geometric locations for mesh vertices by using a mesh smoothing algorithm. In this paper the authors present a new parallel algorithm for mesh smoothing that has a fast parallel runtime both in theory and in practice. The authors present an efficient implementation of the algorithm that uses non-smooth optimization techniques to find the new location of each vertex. Finally, they present experimental results obtained on the IBM SP system demonstrating the efficiency of this approach.
Date: December 31, 1995
Creator: Freitag, L.; Plassmann, P. & Jones, M.
Partner: UNT Libraries Government Documents Department

User guide to power management for PCs and monitors

Description: Power management of personal computers (PCs) and monitors has the potential to save significant amounts of electricity as well as deliver other economic and environmental benefits. The Environmental Protection Agency`s ENERGY STAR{reg_sign} program has transformed the PC market so that equipment capable of power management is now widely available. However, previous studies have found that many Energy Star compliant computer systems are not accomplishing energy savings. The principal reasons for this are systems not being enabled for power management or a circumstance that prevents power management from operating. This guide is intended to provide information to computer support workers to increase the portion of systems that successfully power manage. The guide introduces power management concepts and the variety of benefits that power management can bring. It then explains how the parts of a computer system work together to enter and leave power management states. Several common computer system types are addressed, as well as the complications that networks bring to power management. Detailed instructions for checking and configuring several system types are provided, along with trouble shooting advice. The guide concludes with a discussion of how to purchase Energy Star compliant systems and future directions for power management of PCs and related equipment.
Date: January 1997
Creator: Nordman, B.; Piette, M. A.; Kinney, K. & Webber, C.
Partner: UNT Libraries Government Documents Department

Software verification and validation for commercial statistical packages utilized by the statistical consulting section of SRTC

Description: The purpose of this report is to provide software verification and validation for the statistical packages used by the Statistical Consulting Section (SCS) of the Savannah River Technology Center. The need for this verification and validation stems from the requirements of the Quality Assurance programs that are frequently applicable to the work conducted by SCS. The IBM Personal Computer 300PL and 300XL are both Pentium II based desktops. Therefore the software verification and validation in this report is valid interchangeably between both platforms. As new computing platforms, statistical packages, or revisions to existing packages are reevaluated using these new tools, this report is to be revised to address their verification and validation.
Date: March 22, 2000
Creator: Edwards, T.B.
Partner: UNT Libraries Government Documents Department

CRADA ORNL 91-0046B final report: Assessment of IBM advanced computing architectures

Description: This was a Cooperative Research and Development Agreement (CRADA) with IBM to assess their advanced computer architectures. Over the course of this project three different architectures were evaluated. The POWER/4 RIOS1 based shared memory multiprocessor, the POWER/2 RIOS2 based high performance workstation, and the J30 PowerPC based shared memory multiprocessor. In addition to this hardware several software packages where beta tested for IBM including: ESSO scientific computing library, nv video-conferencing package, Ultimedia multimedia display environment, FORTRAN 90 and C++ compilers, and the AIX 4.1 operating system. Both IBM and ORNL benefited from the research performed in this project and even though access to the POWER/4 computer was delayed several months, all milestones were met.
Date: February 1, 1996
Creator: Geist, G.A.
Partner: UNT Libraries Government Documents Department

A Parallel Genetic Algorithm for the Set Partitioning Problem

Description: In this dissertation the author reports on his efforts to develop a parallel genetic algorithm and apply it to the solution of set partitioning problem -- a difficult combinatorial optimization problem used by many airlines as a mathematical model for flight crew scheduling. He developed a distributed steady-state genetic algorithm in conjunction with a specialized local search heuristic for solving the set partitioning problem. The genetic algorithm is based on an island model where multiple independent subpopulations each run a steady-state genetic algorithm on their subpopulation and occasionally fit strings migrate between the subpopulations. Tests on forty real-world set partitioning problems were carried out on up to 128 nodes of an IBM SP1 parallel computer. The authors found that performance, as measured by the quality of the solution found and the iteration on which it was found, improved as additional subpopulation found and the iteration on which it was found, improved as additional subpopulations were added to the computation. With larger numbers of subpopulations the genetic algorithm was regularly able to find the optimal solution to problems having up to a few thousand integer variables. In two cases, high-quality integer feasible solutions were found for problems with 36,699 and 43,749 integer variables, respectively. A notable limitation they found was the difficulty solving problems with many constraints.
Date: May 1994
Creator: Levine, David
Partner: UNT Libraries Government Documents Department

Elemental ABAREX -- a user's manual.

Description: ELEMENTAL ABAREX is an extended version of the spherical optical-statistical model code ABAREX, designed for the interpretation of neutron interactions with elemental targets consisting of up to ten isotopes. The contributions from each of the isotopes of the element are explicitly dealt with, and combined for comparison with the elemental observables. Calculations and statistical fitting of experimental data are considered. The code is written in FORTRAN-77 and arranged for use on the IBM-compatible personal computer (PC), but it should operate effectively on a number of other systems, particularly VAX/VMS and IBM work stations. Effort is taken to make the code user friendly. With this document a reasonably skilled individual should become fluent with the use of the code in a brief period of time.
Date: May 26, 1999
Creator: Smith, A.B.
Partner: UNT Libraries Government Documents Department

IBM SP high-performance networking with a GRF.

Description: Increasing use of highly distributed applications, demand for faster data exchange, and highly parallel applications can push the limits of conventional external networking for IBM SP sites. In technical computing applications we have observed a growing use of a pipeline of hosts and networks collaborating to collect, process, and visualize large amounts of realtime data. The GRF, a high-performance IP switch from Ascend and IBM, is the first backbone network switch to offer a media card that can directly connect to an SP Switch. This enables switch attached hosts in an SP complex to communicate at near SP Switch speeds with other GRF attached hosts and networks.
Date: May 27, 1999
Creator: Navarro, J.P.
Partner: UNT Libraries Government Documents Department

RJESTATS: a system of programs for remote job entry station accounting

Description: The RJESTATS system of programs provides accounting information on the use of the Environmental Sciences Division Remote Job Entry (RJE) facility. Summaries can be generated to display the number and kind of users, types of jobs run through the RJE facility, and the persons (or projects) that are utilizing the facility. Sample tables are presented to show certain summaries by user name, job name initials, and types of jobs run. RJESTATS is useful for managing and determining the use of an IBM-compatible RJE facility.
Date: March 1, 1978
Creator: Strand, R. H.; Raine, G. B.; Tharp, M. L. & Griffith, N. A.
Partner: UNT Libraries Government Documents Department

Users guide for the ANL IBM SPx

Description: This guide presents the features of the IBM SPx installed in the Mathematics and Computer Science Division at Argonne National Laboratory. The guide describes the available hardware and software, access policies, and hints for using the system productively.
Date: December 1, 1994
Creator: Gropp, W. & Lusk, E.
Partner: UNT Libraries Government Documents Department

Comparison of the AMDAHL 470V/6 and the IBM 370/195 using benchmarks

Description: Six groups of jobs were run on the IBM 370/195 at the Applied Mathematics Division (AMD) of Argonne National Laboratory using the current production versions of OS/MVT 21.7 and ASP 3.1. The same jobs were then run on an AMDAHL 470V/6 at the AMDAHL manufacturing facilities in Sunnyvale, California, using the identical operating systems. Performances of the two machines are compared. Differences in the configurations were minimized. The memory size on each machine was the same, all software which had an impact on run times was the same, and the I/O configurations were as similar as possible. This allowed the comparison to be based on the relative performance of the two CPU's. As part of the studies preliminary to the acquisition of the IBM 195 in 1972, two of the groups of jobs had been run on a CDC 7600 by CDC personnel in Arden Hills, Minnesota, on an IBM 360/195 by IBM personnel in Poughkeepsie, New York, and on the AMD 360/50/75 production system in June, 1971. 6 figures, 9 tables.
Date: March 1, 1976
Creator: Snider, D. R.; Midlock, J. L.; Hinds, A. R. & Engert, D. E.
Partner: UNT Libraries Government Documents Department

Fast exponential calculation for the IBM 370-195 at the Oak Ridge Gaseous Diffusion Plant

Description: The time spent for computing an exponential for the IBM 370-195 at the Oak Ridge Gaseous Diffusion Plant is about 25 times slower than that spent by the CDC-7600 CTR computer at Livermore. We propose a scheme which does not involve a function call, is 15 times faster than the IBM 370-195 scheme, and is useful when the exponential is called many times.
Date: January 1, 1977
Creator: Whitson, J. C. & Whealton, J. H.
Partner: UNT Libraries Government Documents Department

Set Theoretic Data Structures (STDS): a tutorial

Description: Extended Set Theory as a data base management discipline has received attention in the data base literature. Set Theoretic Information Systems Corporation has, for some time, marketed a data base system based on the foundation of Extended Set Theory. This system is called Set Theoretic Data Structures (STDS). A series of examples shows that STDS is similar to relational algebraic data base management systems. The advantages of STDS are its straight-forward data base design, compact data representation, and flexible, powerful data manipulation operators, while its limitations are its low-level primitive user interface and the partial implementation of the Extended Set Theoretic concepts. To make STDS very attractive, a ''user-friendly'' interface should be developed, and some distinctive features of extended Set Theory (such as sets of sets) should be implemented. 8 figures, 1 table.
Date: January 31, 1977
Creator: Birss, E. W. & Yeh, J. W.
Partner: UNT Libraries Government Documents Department

KENO IV: an improved Monte Carlo criticality program

Description: KENO IV is a multigroup Monte Carlo criticality program written for the IBM 360 computers. It executes rapidly and is flexibly dimensioned so the allowed size of a problem (i.e., the number of energy groups, number of geometry cards, etc., are arbitrary) is limited only by the total data storage required. The input data, with the exception of cross sections, fission spectra and albedos, may be entered in free form. The geometry input is quite simple to prepare and complicated three-dimensional systems can often be described with a minimum of effort. The results calculated by KENO IV include k-effective, lifetime and generation time, energy-dependent leakages and absorptions, energy- and region-dependent fluxes and region-dependent fission densities. Criticality searches can be made on unit dimensions or on the number of units in an array. A summary of the theory utilized by KENO IV, a section describing the logical program flow, a compilation of the error messages printed by the code and a comprehensive data guide for preparing input to the code are presented. 14 references. (auth)
Date: November 1, 1975
Creator: Petrie, L.M. & Cross, N.F.
Partner: UNT Libraries Government Documents Department

ETOE-2/MC$sup 2$-2/SDX multigroup neutron cross-section processing

Description: The ETOE-2/MC/sup 2/-2/SDX code system provides a comprehensive neutron cross-section processing capability. The three components of the system include an ENDF/B library processor (ETOE-21, an ultra-fine-group fundamental mode calculation (MC/sup 2/-2), and an intermediate-group space-dependent capability (SDX). Brief descriptions of the types of calculations are discussed. Selected results from typical problem executions are presented. (auth)
Date: January 1, 1973
Creator: Henryson, H. II; Toppel, B.J. & Stenberg, C.G.
Partner: UNT Libraries Government Documents Department