Externally Funded Research ProjectsParallel and Distributed Computing
Externally Funded Research Projects
MATCH: A MATLAB Compilation Environment for Adaptive Computing Systems
Investigators: P. Banerjee, A. Choudhary, S. Hauck, and N. Shenoy
Sponsor: Defense Advanced Research Projects Agency (DARPA), 4/98 – 3/01
Adaptive computing systems constitute a new class of computing and communication technology composed of configurable hardware capable of system-level adaptation. Such systems are often built out of combinations of microprocessor-based embedded systems, specialized digital signal processors (DSP’s), and field-programmable gate arrays (FPGA’s). The objective of the MATCH (MATLAB Compiler for Heterogeneous adaptive computing systems) project is to make it easier for users to develop efficient codes for adaptive computing systems. As part of this project, we are developing a compiler that allows input of a user’s applications written in the high-level language, MATLAB, and generates efficient low-level code that runs on commercial off-the-shelf FPGA’s, embedded processors, and DSP’s. Our specific aims include:
• development of a hardware testbed consisting of commercial FPGA’s,
embedded processors, and DSP’s;
• development of a basic compiler for mapping a given MATLAB
application onto this heterogeneous target;
• investigation of automated parallelization and mapping techniques;
• design and support of compiler directives;
• development of library functions and applications of interest to DOD;
• development of faster algorithms for compilation.
PANTHER: A High-Performance Distributed Computing Infrastructure
Investigators: P. Banerjee, A. Choudhary, S. Hauck, D. T. Lee, M. Sarrafzadeh, P. Scheuermann, and V. Taylor
Sponsor: National Science Foundation (NSF), 9/97 – 8/02
Specific aims of the PANTHER project include:
file-systems and data-management issues for high-performance
distributed computing;
heterogeneous environment;
advantage of high-speed networking;
bandwidth and high speed.
This project provides for the acquisition of 50 high-performance Hewlett-Packard C-180 UNIX workstations, 20 medium-performance B-132 UNIX workstations; 3 J-280 UNIX fileservers, an 8-processor Silicon Graphics Origin 2000 distributed shared-memory multiprocessor, and four CISCO systems LS1010 ATM switches. All of the machines are connected via an OC-3 ATM network operating at 155 Mbps.
PARADIGM: Efficient Compilation Issues in Distributed Memory Multicomputers
Principal Investigator: P. Banerjee
Sponsor: NSF, 9/96 – 9/99
In this research, we are developing efficient compilation and run-time techniques for optimized code generation for distributed-memory multicomputers and distributed shared-memory multiprocessors. We are producing the next generation PARADIGM compiler to handle general block, cyclic and block-cyclic data distributions in regular applications. We are also developing techniques using global dataflow analysis to perform redundant array access elimination. We are investigating techniques for handling regular and irregular codes in a uniform framework based on a runtime library.
Filtration Combustion for Microgravity Applications: (1) Smoldering; (2) Combustion Synthesis of Advanced Materials
Investigators: A. Bayliss, B. J. Matkowsky, and V. Volpert (Applied Mathematics)
Sponsor: National Aeronautics and Space Administration (NASA), 6/1/94 – 11/30/98
In this grant, we study filtration combustion with an emphasis on combustion synthesis and smoldering. The objectives are to understand these processes, describe observed modes of combustion, and predict new modes. The investigations employ both analytical and numerical methods. We consider a variety of physical effects including cellular flames in filtration combustion, and the dynamics of solid-fuel combustion when melting occurs.
Investigation of Nuclear Burning on Neutron Stars
Investigators: A. Bayliss and R. Taam (Physics)
Sponsor: NASA , 2/15/98 – 2/14/99
We investigate the role of nuclear burning on accreting neutron stars in the X-ray burst phenomenon. Particular attention is focused on the study of instabilities of the burning front and its evolution to the previously unexplored pulsing regime. The ignition, nuclear evolution, and propagation of symmetric and asymmetric conductive burning fronts on the surface of a neutron star in multi-dimensional circumstances is investigated. Studies also include a multi-dimensional analysis of a nuclear burning front in which energy is transported by convection. We provide detailed treatments of the nuclear physics, thermodynamics, and gas dynamics using hydrodynamic, nuclear reaction network, and flame-propagation computer codes based on adaptive spectral and finite-difference methods that we previously developed. The results are used to determine the propagation speed and variability of the conductive and convective burning fronts, thereby providing fundamental understanding of nuclear burning on the surfaces of accreting neutron stars. The resulting influence of multi-dimensional effects on the X-ray light curve offers additional insights in interpreting the bursting behavior and variability seen in observed sources. Armed with these new theoretical models and the prospect of new observations with the Rossi X-Ray Timing Explorer, the potential for new insights into the X-ray burst phenomenon is high.
Non-Reflecting Boundary Conditions Based on Far Field Expansions
Principal Investigator: A. Bayliss
Sponsor: NSF, 7/96 – 6/99
In this project we develop, implement, and assess outer grid boundary conditions for wave-propagation problems that are based on far-field expansions of the solution. The methodology is to obtain a far-field expansion for the solution, for example, in terms of the distance from a given source location, and then derive outer boundary conditions as differential operators that annihilate progressively more terms in the expansion. Techniques of interest use interior information to improve the performance of the outer boundary conditions. This permits dealing with situations where the source location is not known in advance, where sources move, or where waves from different sources impinge on different portions of the boundary at the same time. We also develop techniques to increase the order of the outer boundary condition without increasing the order of the spatial derivatives in the boundary operator. This is done by incorporating inhomogeneous terms in the boundary condition derived from far-field expansions of the solution.
Compiler and Run-Time Optimization Techniques for Parallel Programming
Principal Investigator: A. Choudhary
Sponsor: NSF Young Investigator Award, 5/93 – 8/00
This project develops compiler and run-time optimization techniques for scalable parallel programming. In particular, the objectives are to develop compiler techniques for parallel I/O and locality optimizations, and run-time techniques for data-redistribution and memory-hierarchy optimizations. Further, this research involves the development of fundamental models and compilation techniques for out-of-core computations.
System Software for Input-Output in Parallel Computers
Principal Investigator: A. Choudhary
Sponsor: NSF, 5/96 – 4/99
Large-scale parallelism and efficient supporting software can mitigate the performance difference between disk systems and CPU’s / interconnects to achieve an overall balanced, scalable operation. However, the use of parallelism in I/O systems introduces complexities for the system software including run-time systems and file systems. This project addresses the system-software problem, specifically run-time support and interface issues, to perform efficient I/O on parallel computers. Project goals include the design and development of a sophisticated run-time system (PASSION) that (1) incorporates the concepts of collective I/O, data reuse and prefetching, and other optimizations; and (2) implements high-level interfaces to facilitate I/O as easily as accessing any data structures from within a program. This approach should diminish I/O bottlenecks and permit the efficient performance of parallel I/O from application programs.
Interoperable Data Files for High-Performance Computing
Principal Investigator: A. Choudhary
Sponsor: NSF / University of Wyoming, 7/97 – 6/99
This project is developing data-management techniques that enhance both interoperability and high-performance parallel access to scientific data. Specific objectives include the development of software support for: (1) "portable" data files that provide user-defined abstractions; (2) automated filtering and conversion techniques capable of extracting the "meaning" of a data file and presenting it in a form compatible with a host application; (3) improved access techniques that permit parallel access to the files while preserving the basic abstractions; (4) run-time techniques that incorporate collective I/O, data reuse, and prefetching strategies; and (5) legacy files.
Design, Development, Benchmarking and Evaluation of Parallel Applications
Principal Investigator: A. Choudhary
Sponsor: U.S. Air Force Systems Command Rome Laboratory, 12/96 – 5/99
This project has the following design and implementation goals: (1) I/O, data-distribution, and task-scheduling techniques for individual components as well as integrated systems in applications such as space-time adaptive processing, fusion, and target detection; (2) I/O techniques for embedded high-performance system applications; (3) data-distribution and redistribution strategies for inter-task and intra-task data communication in real-time pipelined parallel applications; (4) task assignment and scheduling techniques to satisfy latency and throughput requirements; and (5) portable parallel software having well-defined interfaces.
Design, Development, Benchmarking and Evaluation of Parallel Algorithms for Signal Processing
Principal Investigator: A. Choudhary
Sponsor: U.S. Air Force Systems Command Rome Laboratory, 12/96 – 5/99
This project entails the design, implementation and evaluation of computationally intensive signal processing applications on high-performance parallel embedded systems. As part of this project, we have developed and deployed techniques for parallelization, task mapping and allocation, parallel pipelined communication, data redistribution for applications consisting of several tasks. Another important goal of this project is to achieve a balance of throughput and latency by using the finite computational resources optimally.
We have completed an implementation of the Rome Labs PRI-Staggered Space-Time Adaptive Processing (STAP) application. This STAP algorithm involves (1) Doppler filter processing, (2) weight computation, (3) beam forming, (4) pulse compression, and (5) CFAR detection. We designed a parallel pipelined implementation that is portable across different parallel machines.
Scalable I/O Initiative
Principal Investigator: A. Choudhary
Sponsor: DARPA / California Institute of Technology, 7/96 – 9/99
The I/O performance of massively parallel processors has not kept pace with their processing and communications capabilities. Poor I/O can severely degrade overall throughput. The need for high-performance I/O is so significant that almost all current parallel computers provide hardware and software support for parallel I/O. This project attacks the I/O problem from a language, compiler, and run-time-support point of view. PASSION software support is targeted for I/O-intensive, loosely synchronous problems. The PASSION Runtime Library provides routines to efficiently perform the I/O required in parallel programs. Another goal of the PASSION compiler is to translate out-of-core programs written in a data-parallel language to node programs with calls to the PASSION library.
High-Performance Data-Management, Access, and Storage Techniques for Tera-Scale Scientific Applications
Investigators: A. Choudhary, P. Banerjee, and V. Taylor
Sponsor: Department of Energy (DOE) ASCI Level-2 Grant
Emerging large-scale scientific experiments and simulations require the storage, management, efficient access, and analysis of hundreds of gigabytes to hundreds of terabytes of data. Current data-management and analysis techniques do not satisfy these needs in term of performance, scalability, ease of use, and interfaces. This project is developing a scalable, high-performance, data-management system (SHPDM) to provide support for data management, query capability, and high-performance accesses to large datasets stored in hierarchical storage systems such as HPSS. SHPDM will provide the flexibility of conventional databases for indexing, searches, management of objects, and creating and keeping histories and trails of data accesses. It will also provide high-performance access methods and optimizations (pooled striping, prefetching, caching, collective I/O) for accessing large-scale data-objects found in scientific computations.
Automated Reasoning Tools for Verification and Testing
Principal Investigator: L. Henschen
Sponsor: Motorola, Inc., 1/99-8/99
A verification system for telephony software is being developed based on the ACL2 system. As ACL2 is both a formal language for representing assertions about software and an executable language, our system can both prove results about the software being modelled and execute that software. Our system will have a library of techniques for telephony software verification. A user interface will allow arbitrary mixing of ACL's descriptions of functions with actual C imlementations of those functions to form an executable module. The verified ACL2 representation will serve as an orcale for generating correct output against which to compare C-module results.
Numerical Optimization
Principal Investigator: J. Nocedal
Sponsor: NSF, 9/96 – 9/98
Interior-point methods have revolutionized linear programming. This project studies how the underlying ideas can be used to design new robust and efficient algorithms for large-scale nonlinear optimization. Part of the study is devoted to the development of the NITRO software which combines new ideas with proven techniques such as trust-region methods and sequential quadratic programming. The investigations include theory, algorithm design, and software development.
Challenges in Cise: Metacomputing Environments for Optimization
Principal Investigator: J. Nocedal
Sponsor: NSF, 9/97 – 8/00
The metacomputing paradigm has emerged as an economical way to harness the power of large, distributed collections of computers, making use of compute cycles that would otherwise be wasted. The goal of this project is to use metacomputing platforms to enable the solution of very large optimization problems that arise in science, engineering, and economics. By combining distributed hardware, network hardware / software infrastructure, and optimization algorithms and modeling tools, we aim to produce environments powerful enough to tackle optimization problems of unprecedented size and complexity.
Interactive Environment for Optimization
Principal Investigator: J. Nocedal
Sponsor: Sandia National Laboratory, 6/98 – 9/98
Our long-term goal is to create a highly interactive mode of communication for nonlinear programming using NEOS Internet-based facilities. The objective of this project is to allow a user to perform all the simulations on his machine and only delegate to NEOS the task of computing a new iterate. This new environment will be especially useful for scientists having very large and confidential simulation codes. Our principal collaborators are Sandia National Laboratories for protein-folding applications and NASA Langley for shape-design optimization.
Large Scale Optimization and its Application to Weather Forecasting
Principal Investigator: J. Nocedal
Sponsor: DOE, 8/95 – 9/98
Modern weather-forecasting techniques make use of the variational assimilation of data. This is an optimization-based method for estimating the unknown conditions of the atmosphere. The number of variables in these simulations is in the order of millions. In this project, we develop new algorithms that can solve problems of this type within the time limits required by operational forecasting. The new optimization techniques are also applicable to climate simulations.
Climate Modeling on Parallel Machines
Principal Investigator: J. Nocedal
Sponsor: Argonne National Laboratory, 4/97 – 9/98
The focus of this project has changed from climate modeling to mixed-integer nonlinear programming using highly efficient branch-and-bound methods. This project combines these techniques with recent developments in nonlinear optimization and emerging computer architectures to tackle very large problems. The new algorithms should have many applications in industry from scheduling to portfolio management.
Nonlinear Programming: Algorithms and Software Environments
Principal Investigator: J. Nocedal
Sponsor: DOE, 8/98-7/01
The projects range from the design of new algorithms and the study of their convergence properties, to the development of production-level software. One of the main goals is to devise new interior point methods for large-scale nonlinear programming. Much attention will also be given to the study of new automatic preconditioners for the conjugate gradient method, and to their application in both optimization calculations and finite element models.
Scalable and Adaptive File Management in Networks of Workstations
Principal Investigator: P. Scheuermann
Sponsor: NASA, 9/96 – 3/00
This project is developing a prototype system to provide a scalable architecture for replication on the Web. The new client / server architecture allows for user-transparent geographical replication of Web services. This system is legacy-friendly, i.e., it requires no changes to the existing Web infrastructure. The clients are downloaded as applets to commercially available browsers, while the Web servers are extended with servlets. Both server and client machines take active roles: server machines determine the resources to be replicated / migrated and the best locations for these resources; while client machines choose the best server to submit their requests.
Putting Log Data to Work: Mass Storage Information Systems
Principal Investigator: P. Scheuermann
Sponsor: NASA, 1/98 –5/00
This study is part of the Mass Storage System Performance Analysis project at NASA Goddard. The goal is to capture useful information about the activity of the mass data storage and delivery system from system logs. Currently, these logs are large, ill-structured, and inaccessible for querying. Therefore, the overall performance of the mass-storage system, as well as the statistics about user access patterns, cannot be easily determined. We are investigating the use of data-warehousing and data-mining techniques to obtain easy access to summaries of the log data, as well as to search for interesting patterns of user activity in the storage system.
SCALAR : Scalable Architecture for Replication on the Web
Principal Investigator: P. Scheuermann
Sponsor: NSF, 8/99-7/02
The wide use of the World-Wide Web and the ease of access to the Internet are changing drastically our requirements in distributed systems. The exponentially growing scale of use of the Web is stressing the capacity of the Internet and leads to poor performance and low reliability of Web service. Replication is one of the major techniques that have been proposed to overcome these problems. This project will focus on the development of a prototype system that will provide a scalable architecture for replication on the Web. A new client/server architecture that will allow for user-transparent geographic replication is investigated. Unlike other proposed Web architectures, both clients and servers take active roles in this architecture. Server machines determine the documents to be replicated/migrated and best locations for the documents, while client machines choose the best server to submit their requests based on estimates of the servers’ prior response times. Heuristic algorithms based on clustering techniques will be developed in order to determine sets of documents that be beneficially migrated/replicated as a unit.
Prophecy: A Hierarchical Tool for Modeling and Analyzing Parallel Scientific Applications
Principal Investigator: V. Taylor
Sponsor: NASA, 1998 – 2000
Currently, there exists a large gap between the theoretical peak performance and the actual achieved performance of advanced parallel computers when running large scientific simulations. This gap stems in part from an incomplete understanding of how the parallel-system features impact the performance of the applications. Detailed performance models aid in understanding the relationship between the parallel system and the application. This project studies the use of several performance models, including nonlinear models, to aid in understanding and predicting application performance.
Young Investigator Award
Principal Investigator: V. Taylor
Sponsor: NSF, 8/93 – 7/99
This grant supports the PI’s research in three key areas related to the performance of parallel scientific applications: (1) development of the mesh-partitioning tool, "PART," for distributed systems; (2) investigation of methods to improve the performance of parallel shortest-path applications; and (3) development of a tool for setting up performance models. The third area is still in an initial stage.
Symmetric Multiprocessors for Research on Parallel Architectures, Compilers, Applications, and On-Demand Network Computing Research
Investigators: V. Taylor, J. Fortes, and R. Eigenmann (Purdue)
Sponsor: NSF, 12/96 – 11/98
The focus of this project is on advancing the state-of-the-art in high-performance computing on four fronts: (1) design of high-performance computers capable of 100-trillion operations per second (100 TOPS); (2) advanced tools for analysis and compilation of programs for execution on machines with multiple processors; (3) characterization of applications that require computation rates of 100–1000 TOPS; and (4) design of a network-based software infrastructure that allows the use of linked computers and linked programs (i.e., metacomputers and metaprograms) through conventional network browsers. The main ideas currently under investigation are being validated through extensive computational experiments and computer simulations. The equipment acquired with the funds provided by this grant provides the computational speed, memory, storage, and communication demands of the experimental component of this project which would otherwise not be possible or take unacceptably long times to perform. This work is a collaboration between researchers at Northwestern University and Purdue University.
A Systems Characterization of Large Computational Applications
Investigators: V. Taylor, J. Fortes, R. Eigenmann, and G. Fox (Purdue)
Sponsor: NSF, 1997 – 98
The characterization of large computational applications from a systems perspective is critical to the design of future high-performance computer systems. There is a growing acceptance of the fact that systems research needs to be driven by realistic applications. This project completes initial work aimed at: (1) developing an infrastructure for systems characterization of applications; and (2) conducting initial characterizations of three scientific applications.
Prophesy: A Performance Network for Developing a Computer Algebra
Investigators: Valerie Taylor and Rick Stevens (University of Chicago and Argonne National Laboratory)
Sponsor: NSF, 8/99 - 7/02
Currently, a large gap exists between the peak performance of high-performance systems and the achieved performance when running large scientific simulations. The large performance gap stems partly from a poor understanding of realistic, large applications and the lack of understanding of how the computer system features impact the performance of the applications. The proposed work addresses this problem by developing a framework, called Prophesy, that facilitates the development of very-fine to coarse-grain analytical performance models. Further, Prophesy will enable the development of an algebra that identifies how elementary performance models of system components and application algorithms should be composed to reflect the performance of the application executing on a parallel system, grid or GiB.
Integration of Computer Architecture and Parallel Programming Tools Into Computer Science and Engineering
Investigators: Valerie Taylor, Alok Choudhary ( Northwestern University), Jose Fortes and Rudolf Eigenmann (Purdue University), Louis Vidal and Jan Jo Chen ( Chicago State University)
Sponsor: NSF, 9/98 - 8/01
The goal of this project is to integrate computer architecture and parallel programming tools into computer science and engineering curricula. We do this using a network computing infrastructure called the Purdue University Network Computing Hubs (PUNCH). Existing courses are being enriched with both content and experiments not previously possible. Three institutions (Purdue University, Northwestern University and Chicago State University) are involved in the project. Northwestern University is involved with integrating several tools into two computer architecture courses. All three universities are developing educational materials and are installing tools for shared use through PUNCH.
Alliance Team B: Distributed Computing
Investigator: Valerie Taylor
Sponsor: NSF (subcontract from NCSA), 10/98-9/99
This subcontract is part of an NSF PACI award to NCSA, called the Alliance. The particular project entails the completion of two major tasks: (1) the development of performance models for the cosmology application executed on distributed systems and (2) the integration of these models into the PART tool, which performs data partitioning for distributed systems. The initial focus will be on the cosmology application because it builds upon the expertise of the researchers at Northwestern University in the area of mesh-based applications.
Design of Assembly Operations for Recycling Disassembly
Investigators: C. Wu and D. T. Lee (Northwestern), S. M. Mok, D. Hong and Thomas Babin, (Motorola)
Sponsor: Motorola, Inc, 1/99-8/99
The process of automatic assembly and disassembly has become a key technology in recycling for protecting our environment. By having automatic disassembly processes to de-manufacture products at a lower cost, recycling efforts can be accelerated and made to be profitable. In collaboration with engineers of the Advanced Technology Center at Motorola, the first phase of our proposal presented a Computer Aided Design and Computer Aided Manufacturing (CAD/CAM) framework of software design and manufacturing tools for assembly and disassembly. The proposed approach broke the product life into different cycles. It starts from the design phase to tooling and manufacturing. The product is then sold and collected for de-manufacturing to complete the cycle. The proposed research is to develop automatic assembly operations for recycling disassembly at Motorola. During the first-phase of studying this problem, a Virtual Assembly and Disassembly (VIRAD) system that can be integrated into a CAD/CAM system is developed for designers to evaluate products for assembly and disassembly efficiencies during the design process. The framework of the proposed system divides a product's life cycle into three groups: a CAD design system, a VIRAD environment, and manufacturing and de-manufacturing processes. In this proposal we will concentrate on designing the VIRAD system.
Book Sections and Chapters
J. Nocedal and S. Wright, Numerical Optimization , Springer Verlag, 650 pp, 1999.
Journal Papers*
* (For ease of identification, each citation in this and the following sections will begin with the group faculty member(s)’ name(s))
P. Banerjee, A. Choudhary, M. Kandemir, N. Shenoy, and J. Ramanujam, "A Linear Algebra Framework for Automatic Determination of Optimal Data Layouts," IEEE Transactions on Parallel and Distributed Systems, Vol. 10, No. 2, Feb. 1999.
A. Bayliss, C. C. Fenno, Jr. and L. Maestrello, "Interaction of Sound from Supersonic Jets with Nearby Structures", AIAA J., Vol. 35, pp. 2153-2169, 1998.
A. Bayliss and B. J. Matkowsky, "Interaction of Counterpropagating Hot Spots in Solid Fuel Combustion," Physica D, Vol. 128, pp. 18-40, 1999.
A. Bayliss, A. Golovin, B. J. Matkowsky and A. A. Nepomnyaschy, "Coupled KS-CGL and Coupled Burgers-CGL Equations for Flames Governed by a Sequential Reaction," Physica D, Vol. 129 pp. 253-298, 1999.
A. Bayliss, "Localized Flickering Cellular Flames," SIAM J. Applied Mathematics, Vol. 59, pp. 942-964, 1999.
A. Choudhary, M. Kandemir and J. Ramanujam, "Improving Cache Locality by a Combination of Loop and Data Transformations," IEEE Transactions on Computers, Vol. 48, No. 2, Feb. 1999.
A.Choudhary, M.Kandaswamy, M. Kandemir, and D. Bernholdt, "An Experimental Study to Analyze and Optimize Hartree-Fock Application’s I/O With PASSION," International Journal of High Performance Computing Applications, Vol. 12, No. 4, pp. 411-439, Winter 1998.
A. Choudhary, P. Banerjee, M. Kandemir, and J. Ramanujam, "An Iteration Space Transformation Algorithm Based on Explicit Data Layout Representation for Optimizing Locality," in Languages and Compilers for Parallel Computing, S. Chatterjee et al.eds, Lecture Notes in Computer Science, Springer-Verlag, 1999.
J. Nocedal, L.T. Biegler, C. Schmidt and D. Ternet, "Numerical Experience with a Reduced Hessian Method for Large-Scale Constrained Optimization," Computational Optimization and Applications , 1999.
J. Nocedal and N.I.M. Gould, "The modified absolute-value factorization norm for trust-region minimization," in High Performance Algorithms and Software in Nonlinear Optimization, R. De Leone, A. Murli, P. M. Pardalos and G. Toraldo, eds., Kluwer, pp. 225-241, 1998.
J. Nocedal and Y.-X. Yuan, "Combining Trust Region and Line Search Techniques," in Advances in Nonlinear Programming, Y. Yuan, ed., Kluwer, pp. 153-175, 1998.
J. Nocedal and R. Byrd, "Active Set and Interior Point Methods for Nonlinear Optimization," Documenta Mathematica, Vol. III, Journal der Deutschen Mathematiker-Vereinigung, 1998.
J. Nocedal, "An Interior Point Method for Large-Scale Nonlinear Programming," SIAM. J. of Optimization, Vol. 9, No. 4, pp. 877-900, 1999.
P. Scheuermann, Junho Shim and Radek Vingralek, " Proxy Cache Design: Algorithms, Implementation and Performance," IEEE Transactions on Knowledge and Data Engineering, (Special issue on Web Technologies), Volume 11, No. 4 , pp. 549-562, July/August 1999.
P. Scheuermann, Peter Zabback, Ibrahim Onyuksel and Gerhard Weikum, "Database Reorganization in Parallel Disk Arrays with I/O Service Stealing," IEEE Transactions on Data and Knowledge Engineering, Volume 11, No. 4, pp. 855-858, September/October 1998.
V.Taylor, M. Hribar, and D. E. Boyce, "Termination Detection for Parallel Shortest Path Algorithms," Journal of Parallel and Distributed Computing, Vol. 55, pp. 153-165, 1998.
V.Taylor, Z. Ben Miled, J. A. B. Fortes, and R. Eigenman "On the Cost-Efficiency of Hierarchical Heterogeneous Machines for Compiler and Hand Parallelized Applications," International Journal of Parallel and Distributed Systems and Networks, Vol. 1:4, pp. 193--203, 1998.
V. Taylor, W. Smith and I. Foster, "Predicting Application Run Times Using Historical Information," Lecture Notes in Computer Science, Vol. 1459, pp. 122-142, 1998.
C. H. Wu, J. Jiang, S. Tang, M. Dalal, and D. G. Hanson, "Integrated Analyzer and Classifier of Glottographic Signals", IEEE Transactions on Rehabilitation Engineering, Vol.6, No.2, pp 227-234, June 1998.
Symposium Papers
P. Banerjee and A. Mishra, "An Algorithm Based Error Detection Scheme for the Multigrid Method," Proc. of the 1999 International Symposium on Fault Tolerant Computing, Madison, WI, June 15-18, 1999.
P. Banerjee, A. Choudhary, M. Kandemir, J. Ramanujam, and E. Ayguade, "An ILP Approach for Optimizing Cache Locality," Proc. 1999 ACM International Conference on Supercomputing (ICS’99), Rhodes, GREECE, June 1999.
P. Banerjee, A. Choudhary, M. Kandemir, and J. Ramanujam, "Improving Locality Using Loop and Data Transformations in an Integrated Framework," Proc. of the 31st International Symposium on Micro-Architecture (MICRO-31), Dallas, TX, Dec. 1998.
P. Banerjee, A. Choudhary, M. Kandemir, J. Ramanujam, and N. Shenoy "Enhancing Spatial Locality Using Data Layout Optimizations," Proc. Euro-Par’98 (Workshop on Automatic Parallelisation), Southampton, UK, Sept. 1998.
P. Banerjee and P. Joisha, "PARADIGM (version 2.0): A New HPF Compilation System," Proc. of the 1999 International Parallel Processing Symposium (IPPS’99), San Juan, Puerto Rico, April 1999.
P. Banerjee and D. Chakrabarti, "A Novel Compilation Framework for Supporting Semi-Regular Distributions in Hybrid Applications," Proc. of the 1999 International Parallel Processing Symposium (IPPS’99), San Juan, Puerto Rico, April 1999.
P. Banerjee and A. Mishra, "A Fault Tolerant Multi-Grid Algorithm," Proc. of the Conference on Parallel and Distributed Computing Systems (PDCS’98), Chicago, IL, Sept. 1998.
A. Choudhary, V. Taylor, M. Kandemir, H. Nagesh, J. No, X. Shen, , S. More, and R. Thakur, "Data Management for Large-Scale Scientific Computations in High Performance Distributed Systems," Proc. 8th IEEE International Symposium on High Performance Distributed Computing, Redondo Beach, CA, Aug. 1999.
A. Choudhary and S. Goil, "A Parallel Scalable Infrastructure for OLAP and Data Mining," Proc.1999 International Database Engineering and Applications Symposium (IDEAS’99), Montreal, CANADA, Aug. 1999.
A. Choudhary, M. Kandemir and J. Ramanujam, "Restructuring I/O-Intensive Computations for Locality," Proc. 7th International Conference on High-Performance Computing and Networking (HPCN Europe 1999), Amsterdam, The Netherlands, April 1999.
A. Choudhary, J. No and J. Carretero, "High Performance Parallel I/O Schemes for Irregular Applications on Clusters of Workstations," Proc. 7th International Conference on High-Performance Computing and Networking (HPCN Europe 1999), Amsterdam, THE NETHERLANDS, April 1999.
A.Choudhary and S. Goil, "An Infrastructure for Scalable Parallel Multidimensional Analysis," Proc.11th International Conference on Scientific and Statistical Databses (SSDBM’99), Cleveland, OH, July 1999.
A. Choudhary, W. Liao, D. Weiner, and P. Varshney, "Multi-Threaded Design and Implementation of Parallel Pipelined STAP on Parallel Computers with SMP Nodes," Proc. 13th International Parallel Processing Symposium, 10th Symposium on Parallel and Distributed Processing (IPPS/SPDP 1999), San Juan, Puerto Rico, April 1999.
A. Choudhary, P. Banerjee, M. Kandemir, and J. Ramanujam, "A Graph Based Framework to Detect Optimal Memory Layouts for Improving Data Locality," Proc. 13th International Parallel Processing Symposium, 10th Symposium on Parallel and Distributed Processing (IPPS/SPDP 1999), San Juan, PUERTO RICO, April 1999.
A. Choudhary and S. Goil, "Design and Implementation of a Scalable Parallel System for Multidimensional Analysis," Proc. 13th International Parallel Processing Symposium & 10th Symposium on Parallel and Distributed Processing (IPPS/SPDP’99), San Juan, PUERTO RICO, April 1999.
A. Choudhary and M. Kandemir, "System-Level Meta-Data for High Performance Data Management," Proc. Third IEEEMeta-Data Conference, Bethesda, MD, April 1999.
A. Choudhary, J. Carretero and J. No, "Optimizing I/O for Irregular Applications on Distributed-Memory Machines," Proc. ACPC’99, Salzburg, AUSTRIA, Feb. 1999.
A. Choudhary, J. No and J. Carretero, "Optimizations to Provide High-Performance Parallel I/O for Irregular Applications," Proc. Applied Informatics ’99, Innsbruck, AUSTRIA, Feb. 1999.
A. Choudhary, and S. More, "Extended Collective I/O for Efficient Retrieval of Large Objects," Proc. 5th International Conference on High Performance Computing (HiPC’98), Chennai, INDIA, Dec. 1998.
A. Choudhary and S. Goil, "High Performance Multidimensional Analysis and Data Mining," Proc. High Performance Networking and Computing Conference (SC’98), Orlando, FL, Nov. 1998.
A. Choudhary and S. Goil, "High Performance Multidimensional Analysis of Large Datasets," Proc. ACM First International Workshop on Data Warehousing and OLAP (DOLAP’98) (in conjunction with CIKM’98), Washington, DC, Nov. 1998.
A. Choudhary, P. Banerjee, M. Kandemir, and J. Ramanujam, "A Matrix-Based Approach to the Global Locality Optimization Problem," Proc. International Conference on Parallel Architectures and Compilation Techniques (PACT’98), Paris, FRANCE, Oct. 1998.
A. Choudhary, J. Carretero, W. Zhu, and X. Shen, "MiPFS: A Multimedia Integrated Parallel File System," International Joint Conference on Information Systems, Raleigh, NC, Oct. 1998.
A. Choudhary, J. Carretero and J. No, "Parallel I/O for Irregular Applications on Distributed Memory Machines," IX Jornadas de Paralelismo, San Sebastian, SPAIN, Sept. 1998.
A.Choudhary, R. Krishnaiyer and Ian Foster, "Performance of a Remote I/O Library in High-PerformanceDistributed Computing Environments," Proc. International Conference on Parallel and Distributed Computing and Systems (PDCS’98), Chicago, IL, Sept. 1998.
W.-C. Lin and C.-C. Lin, "Center-On/Off Receptive Field Operators for Face Detection in Generic Scenes," Proceedings of the 4th Joint Conference on Information Sciences, Vol.4, pp. 291-294, Research Triangle Park, North Carolina, Oct. 23-28, 1998.
W.-C. Lin, P. Chen, and H.-L. Hung, "Multi-Resolution Fuzzy ART Neural Networks," Proceedings of International Joint Conference on Neural Networks, Washington, DC, July 10-16, 1999.
W.-C. Lin, P. Chen, H.-L. Hung and D. Prine, "Multi-Resolution Fuzzy ART Neural Networks for Acoustic Signal Classification," Proceedings of the IASTED International Conference on Artificial Intelligence and Soft Computing, Honolulu, Hawaii, Aug. 9-12, 1999.
P. Scheuermann, Radek Vingralek, Yuri Breitbart and Mehmet Sayal, "Web++: A System for Fast and Reliable Web Service," Proceedings of USENIX 1999 Annual Conference, Monterey, pp. 171-184, June 1999.
P. Scheuermann, Yuri Breitbart, Radek Vingralek and Mehmet Sayal ,"A Transparent Replication of HTTP Service" Proceedings of International Conference on Data Engineering (ICDE), Sydney, Australia, p. 97, March 1999.
P. Scheuermann, L. Singh, B. Chen, and R. Height, "An Algorithm for Constrained Association Rule Mining," Proc. Pacific Asia Knowledge Discovery in Databases Conference, , Beijing, China, pp. 148-158, April 1999.
P. Scheuermann, L. Singh, M. Macie, J. Woytek and J. Pan, "Gaining Knowledge for Ill-Structured Data," Proc. of the AFCEA Federal Data Mining Symposium, McLean, VA., pp. 155-161, March 1999.
P. Scheuermann, Lisa Singh and Bin Chen, "IRIS: Our Prototype Rule Generation System," Proceedings SPIE, Orlando, Florida, 1999.
P. Scheuermann, Junho Shim and Radek Vingralek, "Dynamic Caching of Query Results for Decision Support Systems, " Proceedings 11th Intern. Conf. on Statistical and Scientific Database Management, Cleveland, pp. 254-263, July 1999.
V. Taylor, R. J. O. Figueiredo, J. A. B. Fortes, Z. B. Miled, and R. Eigenmann, "Impact of Computing-in-Memory on the Performance of Processor-and-Memory Hierarchies," Proceedings of International Conference on Parallel and Distributed Computing Systems, September 1998.
V. Taylor and J. Geisler, "Performance Coupling: A Methodology for Predicting Application Performance using Kernel Performance," Proceedings of the 1999 SIAM Conference on Parallel Processing, February 1999.
V. Taylor J. Chen, "Mesh Partitioning for Distributed Systems: Exploring Optimal Number of Partitions with Local and Remote Communication," Proceedings of the 1999 SIAM Conference on Parallel Processing, February 1999.
V. Taylor and J. Chen, "ParaPART: Parallel Mesh Partitioning Tool for Distributed Systems", in Proceedings th International Workshop on Solving Irregularly Structured Problems in Parallel), San Juan, Puerto Rico, April 1999.
V. Taylor W. Smith, and I. Foster, "Using Run-Time Predictions to Estimate Queue Wait Times and Improve Scheduler Performance," Proceedings of the IPPS/SPDP '9 Workshop on Job Scheduling Strategies for Parallel Processing, April 1999.
V. Taylor, J. Fortes and R. Eigenmann, "HPAM Petaflop Point Design: Identifying Critical Research Issues for Petaflop Machines," Proceedings of the PetaFlop Workshop (TPF-3), February 1999.
V.Taylor, J. Fortes, R. Eigenmann, A. Choudhary, L. Vidal, and J. Chen, "Integration of Computer Architecture and Parallel Programming Tools into Computer Science and Engineering Curriculum through Network Computing Hubs," Proceedings of the ASEE Annual Conference, June 1999.
C. H. Wu and J.S. Gyorfi , "Coordinated Planning and Control of Multiple Robots and Machines for Surface Mount Manufacturing", Proceedings of the IEEE International Joint Symposia on Intelligence and Systems, pp. 268-272, 1998.
C. H. Wu , D. T. Lee, S. L. Chang, , "Design of an Active Suspension System Based on Muscle-Like Control," 1999 IEEE International Conference on Robotics and Automation, Detroit, May 1999.
Invited Talks and Seminars
P. Banerjee, " A MATLAB Compilation Environment for Adaptive Computing Systems," Georgia Institute of Technology, Jan. 1999.
P. Banerjee, distinguished lecturer, " A MATLAB Compilation Environment for Adaptive Computing Systems," University of Florida, Jan. 1999.
A. Bayliss, "New Modes of Burning in Solid Fuel Combustion", Nonlinear Dynamics Seminar, Mathematics Department, Northwestern University, 1999.
A. Choudhary, "An Architecture for Large-Scale Data Management" Lawrence Livermore Laboratory, Livermore, CA, Jan. 1999.
A. Choudhary, invited talk , "Issues in High-Performance I/O for Teraflops Computing," HPCN’99, Amsterdam, The Netherlands, April 1999.
A . Choudhary, "PARSIMONY: Parallel and Scalable OLAP and Data Mining," University of Vienna, August 1999.
J. Nocedal, "Interior methods for nonlinear programming," INFORMS Meeting, Seattle, WA, October 1998.
J. Nocedal, "Weather forecasting using numerical optimization," Dept of Statistics and Econometrics, Universidad Carlos III, Spain, April 1999.
J. Nocedal, "Automatic Preconditioners," Courant Institute of Mathematical Sciences at New York University, April 1999.
J. Nocedal, plenary talk, "Some Open Questions in Nonlinear Optimization," Foundations of Computational Mathematics, Oxford, England, July 1999.
J. Nocedal, "Handling ill-conditioned Jacobians, " TC7 Conference, Cambridge, England, July 1999.
P. Scheuermann, "Data Mining for Semi-Structured Data," Hewlett Packard Research Labs, February 1999.
P. Scheuermann, "Data Mining on the Web," Illinois Institute of Technology, March 1999.
V.Taylor, keynote speaker, "To Prosper and not Just Survive: That is the Question," Workshop for Undergraduates in EECS, University of California at Berkeley, September 1998.
V.Taylor, "Performance Coupling: A Methodology for Analyzing Application Performance", NASA Ames Research Center, September 1999.
V.Taylor, "PART: A Partitioning Tool for Efficient Use of Distributed Systems", EE Department, University of California at Berkeley, September 1998.
V.Taylor, "Efficient Use of Distributed Systems for Engineering Design & Prototyping", ICAIRE Press Conference, April 1999.
V.Taylor, "Performance Coupling: A Methodology for Analyzing Application Performance," IBM Workshop on Java Server Performance, May 1999.
Symposium Sessions Organized / Chaired
P. Banerjee, program chair, 6th International Conference on High Performance Computing, Calcutta, INDIA, December 1999.
A. Choudhary, program vice chair, 6th International Conference on High Performance Computing (HiPC’99)
J. Nocedal, session chair, SIAM Conference on Optimization, Atlanta GA, May 1999.