Eric Barszcz

Followers

Following

Co-author

Public Views

Interests

Uploads

Papers by Eric Barszcz

Optimal Design of Integrated Systems Health Management (ISHM) Systems for improving safety in NASA's Exploration Vehicles: A Two-Level Multidisciplinary Design Approach

Integrated Vehicle Health Management (ISHM) systems are used to detect, assess, and isolate funct... more Integrated Vehicle Health Management (ISHM) systems are used to detect, assess, and isolate functional failures in order to improve safety of space systems such as Orbital Space Planes (OSP's). An ISHM system, as a whole, consists of several subsystems that monitor different components of an OSP including2: Spacecraft, Launch Vehicle, Ground Control, and the International Space Station. In this research, therefore, we propose a new methodology to design and optimize ISHM as a distributed system with multiple disciplines (that correspond to literature to the multidisciplinary design optimization of problems with such architecture3 (as will be reviewed in the full paper).

Download

The Nas Parallel Benchmarks

The international journal of supercomputer applications, Sep 1, 1991

A new set of benchmarks has been developed for the performance evaluation of highly parallel supe... more A new set of benchmarks has been developed for the performance evaluation of highly parallel supercomputers. These benchmarks consist of ve \parallel kernel" benchmarks and three \simulated application" benchmarks. Together they mimic the computation and data movement c haracteristics of large scale computational uid dynamics applications. The principal distinguishing feature of these benchmarks is their \pencil and paper" speci cation | all details of these benchmarks are speci ed only algorithmically. In this way m a n y of the di culties associated with conventional benchmarking approaches on highly parallel systems are avoided.

Download

A static data flow simulation study at Ames Research Center

International Conference on Parallel Processing, Jun 1, 1987

Demands in computational power, particularly in the area of computational fluid dynamics (CFD), h... more Demands in computational power, particularly in the area of computational fluid dynamics (CFD), have lead NASA Ames Research Center to study advanced computer architectures. based on research done by Jack B. Dennis at Massachusetts Institute of Technology. To improve understanding of this architecture, a static data flow simulator, written in Pascal, has been implemented for use on a Cray X-MP/48. two-dimensional fast Fourier transform (FFT), two algorithms used in CFD work at Ames Research Center, have been run on the simulator. factor of more than 2 depending on the partitioning method used to assign instructions to processing elements. Service time for matching tokens has proved to be a major bottleneck. the execution time. The best sustained MFLOPS rates were less than 50% of the maximum capability of the machine.

format_quoteLate binding partitioner reduces execution time by an average of 41.7% for fully unrolled matrix multiply matrices compared to hand partitioning.format_quote

Download

Advances in Functional Fault Modeling for the Design and Operation of Space Systems

Advances in Computational Intelligence and Autonomy for Aerospace Systems

The Nas Parallel Benchmarks

The International Journal of Supercomputing Applications, 1991

A new set of benchmarks has been developed for the performance evaluation of highly parallel supe... more A new set of benchmarks has been developed for the performance evaluation of highly parallel supercom puters. These consist of five "parallel kernel" bench marks and three "simulated application" benchmarks. Together they mimic the computation and data move ment characteristics of large-scale computational fluid dynamics applications. The principal distinguishing feature of these benchmarks is their "pencil and paper" specification—all details of these benchmarks are specified only algorithmically. In this way many of the difficulties associated with conventional bench- marking approaches on highly parallel systems are avoided.

Another orthogonal matrix

Linear Algebra and its Applications, 2006

A Householder reflector and a suitable product of Givens rotations are two well known methods for... more A Householder reflector and a suitable product of Givens rotations are two well known methods for generating an orthogonal matrix with a given first column. Based on a careful realization of an observation by Householder and Fox, we present a new representation of an orthogonal Hessenberg matrix. We relate our matrix to Schur parameters.

Download

Performance of a Euler code on hypercubes

ABSTRACT This paper evaluates the performance of hypercube machines on a computational fluid dyna... more ABSTRACT This paper evaluates the performance of hypercube machines on a computational fluid dynamics problem. Our evaluation focuses on a widely used fluid dynamics code, FLO52, written by Antony Jameson, which solves the two-dimensional steady Euler equations describing flow around an airfoil. In this paper, we give a description of FLO52, its hypercube mapping, and the code modifications to increase machine utilization. Results from two hyper-cube computers (a 16 node iPSC/2, and a 512 node NCUBE/ten) are presented and compared. In addition, we develop a mathematical model of the execution time as a function of several machine and algorithm parameters. This model accurately predicts the actual run times obtained. Predictions about future hypercubes are made using this timing model.

Performance of a parallel code for the Euler equations on hypercube computers

The performance of hypercubes were evaluated on a computational fluid dynamics problem and the pa... more The performance of hypercubes were evaluated on a computational fluid dynamics problem and the parallel environment issues were considered that must be addressed, such as algorithm changes, implementation choices, programming effort, and programming environment. The evaluation focuses on a widely used fluid dynamics code, FLO52, which solves the two dimensional steady Euler equations describing flow around the airfoil. The code development experience is described, including interacting with the operating system, utilizing the message-passing communication system, and code modifications necessary to increase parallel efficiency. Results from two hypercube parallel computers (a 16-node iPSC/2, and a 512-node NCUBE/ten) are discussed and compared. In addition, a mathematical model of the execution time was developed as a function of several machine and algorithm parameters. This model accurately predicts the actual run times obtained and is used to explore the performance of the code i...

A Blind Permutation Similarity Algorithm

ArXiv, 2020

This paper introduces a polynomial blind algorithm that determines when two square matrices, $A$ ... more This paper introduces a polynomial blind algorithm that determines when two square matrices, $A$ and $B$, are permutation similar. The shifted and translated matrices $(A+\beta I+\gamma J)$ and $(B+\beta I+\gamma J)$ are used to color the vertices of two square, edge weighted, rook's graphs. Then the orbits are found by repeated symbolic squaring of the vertex colored and edge weighted adjacency matrices. Multisets of the diagonal symbols from non-permutation similar matrices are distinct within a few iterations, typically four or less.

format_quoteA stable partition is reached in six iterations or less, separating non-permutation similar matrices in four iterations or less, supporting algorithm efficiency.format_quote

Download

One Year with an iPSC/SSO

This paper describes experiences over the past year with an the Intel iPSC/860, a distributed mem... more This paper describes experiences over the past year with an the Intel iPSC/860, a distributed memory MIMD parallel computer based on the Intel i860 floating point processor. The system at NASA Ames Research Center has 128 nodes, and a theoretical peak performance of over seven GFLOPS. This paper describes the system at Ames Research Center, talks about system stability, compiler performance measured by the NAS kernels, and results from a two-dimensional computational fluid dynamics application.

FL052 on hypercubes : performance of a parallel code for the Euler equations on hypercubes

Qr factorization revisited

A general framework capturing known QR factorization methods is presented. A relationship between... more A general framework capturing known QR factorization methods is presented. A relationship between the major classes of methods leads to a new, efficient and accurate method for computing the Cholesky decomposition of ( I − qqT). Incorporating the new Cholesky decomposition into an accurate QR factorization algorithm is expressed as a splitting of orthogonal Hessenberg matrices. Computation is further optimized by introducing a new factorization for orthogonal Hessenberg matrices called DST factorization. A QR algorithm based on DST factorization is presented that has accuracy comparable to Householder transformations, the flexibility of Givens rotations and the lowest operation count of known methods. Also, it has provable bounds on the computation. Results from computing the QR factorization of several ill-conditioned and rank deficient matrices are presented.

New loop control structures for static data flow

Dynamic overset grid communication on distributed memory parallel processors

11th Computational Fluid Dynamics Conference, 1993

A parallel distributed memory implementation of intergrid communication for dynamic overset grids... more A parallel distributed memory implementation of intergrid communication for dynamic overset grids is presented. Included are discussions of various options considered during development. Results are presented comparing an Intel iPSC/860 to a single processor Cray Y-MP. Results for grids in relative motion show the iPSC/860 implementation to be faster than the Cray implementation.

Healthwatch-2 System Overview

Healthwatch-2 (HW-2) is a research tool designed to facilitate the development and testing of in-... more Healthwatch-2 (HW-2) is a research tool designed to facilitate the development and testing of in-flight health monitoring algorithms. HW-2 software is written in C/C++ and executes on an x86-based computer running the Linux operating system. The executive module has interfaces for collecting various signal data, such as vibration, torque, tachometer, and GPS. It is designed to perform in-flight time or frequency averaging based on specifications defined in a user-supplied configuration file. Averaged data are then passed to a user-supplied algorithm written as a Matlab function. This allows researchers a convenient method for testing in-flight algorithms. In addition to its in-flight capabilities, HW-2 software is also capable of reading archived flight data and processing it as if collected in-flight. This allows algorithms to be developed and tested in the laboratory before being flown. Currently HW-2 has passed its checkout phase and is collecting data on a Bell OH-58C helicopter...

format_quoteCubic spline methods retain more energy from raw signals than faster methods; accuracy prioritized for future research despite slower compute speeds.format_quote

Download

NAS Parallel Benchmark Results David H. Bailey, Eric Barszcz, Leonardo Dagum and Horst D. Simon January 10, 1993

ABSTRACT The NAS Parallel Benchmarks have been developed at NASA Ames Research Center to study th... more ABSTRACT The NAS Parallel Benchmarks have been developed at NASA Ames Research Center to study the performance of parallel supercomputers. The eight benchmark problems are specified in a &quot;pencil and paper&quot; fashion. In other words, the complete details of the problem to be solved are given in a technical document, and except for a few restrictions, benchmarkers are free to select the language constructs and implementation techniques best suited for a particular system. This paper presents performance results of various systems using the NAS Parallel Benchmarks. These results represent the best results that have been reported to us for the specific systems listed. Bailey and Barszcz are employees of NASA Ames Research Center. Dagum and Simon are employees of Computer Science Corp., and their work was funded by the NASA Ames Research Center under contract NAS2-12961. Mailing address for all authors: NASA Ames Research Center, Mail Stop T045-1, Moffett Field, CA 94035. 1 1 Introduction The Num...

Flo52 on Hypercubes: Performance and Modelling of a Multigrid Euler Code

Ijhsc, 1989

An Exploration of Discontinuous Time Synchronous Averaging for Helicopter HUMS Using Cruise and Terminal Area Vibration Data

Recent research using NASA Ames AH-1 and OH-58C helicopters, and NASA Glenn test rigs, has shown ... more Recent research using NASA Ames AH-1 and OH-58C helicopters, and NASA Glenn test rigs, has shown that in-flight vibration data are typically non-stationary [l-41. The nature and extent of this non-stationarity is most likely produced by several factors operating simultaneously. The aerodynamic flight environment and pilot commands provide continuously changing inputs, with a complex dynamic response that includes automatic feedback control from the engine regulator. It would appear that the combined effects operate primarily through an induced torque profile, which causes concomitant stress modulation at the individual internal gear meshes in the transmission. This notion is supported by several analyses, which show that upwards of 93% of the vibration signal's variance can be explained by knowledge of torque alone. That this relationship is stronger in an AH-1 than an OH-58, where measured non-stationarity is greater, suggests that the overall mass of the vehicle is an importan...

format_quoteFor the AH-1, Q state variable explains 75-86% of the variability in mesh energy, underscoring its dominance in power spectral distribution.format_quote

Download

Solution of Regular Sparse Triangular Systems on Vector and Distributed-Memory Multiprocessors

Healthwatch-2 System Overview

Download

Eric Barszcz

Uploads

Papers by Eric Barszcz

Log In