Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed parallelization”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 595 records · Page 33

Arranging computer architectures to create higher-performance controllers

Techniques for integrating microprocessors, array processors, and other intelligent devices in control systems are reviewed, with an emphasis on the (re)arrangement of components to form distributed or parallel processing systems. Consideration is given to the selection of the host microprocessor, increasing the power and/or memory capacity of the host, multitasking software for the host, array processors to reduce computation time, the allocation of real-time and non-real-time events to different computer subsystems, intelligent devices to share the computational burden for real-time events, and intelligent interfaces to increase communication speeds. The case of a helicopter vibration-suppression and stabilization controller is analyzed as an example, and significant improvements in computation and throughput rates are demonstrated.

Jacklin, Stephen A.↗

Adaptive pattern recognition by mini-max neural networks as a part of an intelligent processor

In this decade and progressing into 21st Century, NASA will have missions including Space Station and the Earth related Planet Sciences. To support these missions, a high degree of sophistication in machine automation and an increasing amount of data processing throughput rate are necessary. Meeting these challenges requires intelligent machines, designed to support the necessary automations in a remote space and hazardous environment. There are two approaches to designing these intelligent machines. One of these is the knowledge-based expert system approach, namely AI. The other is a non-rule approach based on parallel and distributed computing for adaptive fault-tolerances, namely Neural or Natural Intelligence (NI). The union of AI and NI is the solution to the problem stated above. The NI segment of this unit extracts features automatically by applying Cauchy simulated annealing to a mini-max cost energy function. The feature discovered by NI can then be passed to the AI system for future processing, and vice versa. This passing increases reliability, for AI can follow the NI formulated algorithm exactly, and can provide the context knowledge base as the constraints of neurocomputing. The mini-max cost function that solves the unknown feature can furthermore give us a top-down architectural design of neural networks by means of Taylor series expansion of the cost function. A typical mini-max cost function consists of the sample variance of each class in the numerator, and separation of the center of each class in the denominator. Thus, when the total cost energy is minimized, the conflicting goals of intraclass clustering and interclass segregation are achieved simultaneously.

Szu, Harold H.↗

Real-Time MENTAT programming language and architecture

Real-time MENTAT, a programming environment designed to simplify the task of programming real-time applications in distributed and parallel environments, is described. It is based on the same data-driven computation model and object-oriented programming paradigm as MENTAT. It provides an easy-to-use mechanism to exploit parallelism, language constructs for the expression and enforcement of timing constraints, and run-time support for scheduling and exciting real-time programs. The real-time MENTAT programming language is an extended C++. The extensions are added to facilitate automatic detection of data flow and generation of data flow graphs, to express the timing constraints of individual granules of computation, and to provide scheduling directives for the runtime system. A high-level view of the real-time MENTAT system architecture and programming language constructs is provided.

Grimshaw, Andrew S.↗

Multiphase complete exchange on a circuit switched hypercube

On a distributed memory parallel computer, the complete exchange (all-to-all personalized) communication pattern requires each of n processors to send a different block of data to each of the remaining n - 1 processors. This pattern is at the heart of many important algorithms, most notably the matrix transpose. For a circuit switched hypercube of dimension d(n = 2(sup d)), two algorithms for achieving complete exchange are known. These are (1) the Standard Exchange approach that employs d transmissions of size 2(sup d-1) blocks each and is useful for small block sizes, and (2) the Optimal Circuit Switched algorithm that employs 2(sup d) - 1 transmissions of 1 block each and is best for large block sizes. A unified multiphase algorithm is described that includes these two algorithms as special cases. The complete exchange on a hypercube of dimension d and block size m is achieved by carrying out k partial exchange on subcubes of dimension d(sub i) Sigma(sup k)(sub i=1) d(sub i) = d and effective block size m(sub i) = m2(sup d-di). When k = d and all d(sub i) = 1, this corresponds to algorithm (1) above. For the case of k = 1 and d(sub i) = d, this becomes the circuit switched algorithm (2). Changing the subcube dimensions d, varies the effective block size and permits a compromise between the data permutation and block transmission overhead of (1) and the startup overhead of (2). For a hypercube of dimension d, the number of possible combinations of subcubes is p(d), the number of partitions of the integer d. This is an exponential but very slowly growing function and it is feasible over these partitions to discover the best combination for a given message size. The approach was analyzed for, and implemented on, the Intel iPSC-860 circuit switched hypercube. Measurements show good agreement with predictions and demonstrate that the multiphase approach can substantially improve performance for block sizes in the 0 to 160 byte range. This range, which corresponds to 0 to 40 floating point numbers per processor, is commonly encountered in practical numeric applications. The multiphase technique is applicable to all circuit-switched hypercubes that use the common e-cube routing strategy.

Bokhari, Shahid H.↗

Macular Bioaccelerometers on Earth and in Space

Space flight offers the opportunity to study linear bioaccelerometers (vestibular maculas) in the virtual absence of a primary stimulus, gravitational acceleration. Macular research in space is particularly important to NASA because the bioaccelerometers are proving to be weighted neural networks in which information is distributed for parallel processing. Neural networks are plastic and highly adaptive to new environments. Combined morphological-physiological studies of maculas fixed in space and following flight should reveal macular adaptive responses to microgravity, and their time-course. Ground-based research, already begun, using computer-assisted, 3-dimensional reconstruction of macular terminal fields will lead to development of computer models of functioning maculas. This research should continue in conjunction with physiological studies, including work with multichannel electrodes. The results of such a combined effort could usher in a new era in understanding vestibular function on Earth and in space. They can also provide a rational basis for counter-measures to space motion sickness, which may prove troublesome as space voyager encounter new gravitational fields on planets, or must re-adapt to 1 g upon return to earth.

Ross, M. D.↗

Responsive systems - The challenge for the nineties

A concept of responsive computer systems will be introduced. The emerging responsive systems demand fault-tolerant and real-time performance in parallel and distributed computing environments. The design methodologies for fault-tolerant, real time and responsive systems will be presented. Novel techniques of introducing redundancy for improved performance and dependability will be illustrated. The methods of system responsiveness evaluation will be proposed. The issues of determinism, closed and open systems will also be discussed from the perspective of responsive systems design.

Malek, Miroslaw↗

PARTI primitives for unstructured and block structured problems

Described here is a set of primitives (PARTI) developed to efficiently execute unstructured and block structured problems on distributed memory parallel machines. We present experimental data from a 3-D unstructured Euler solver run on the Intel Touchstone Delta to demonstrate the usefulness of our methods.

Sussman, Alan↗

Hyperswitch Communication Network Computer

Hyperswitch Communications Network (HCN) computer is prototype multiple-processor computer being developed. Incorporates improved version of hyperswitch communication network described in "Hyperswitch Network For Hypercube Computer" (NPO-16905). Designed to support high-level software and expansion of itself. HCN computer is message-passing, multiple-instruction/multiple-data computer offering significant advantages over older single-processor and bus-based multiple-processor computers, with respect to price/performance ratio, reliability, availability, and manufacturing. Design of HCN operating-system software provides flexible computing environment accommodating both parallel and distributed processing. Also achieves balance among following competing factors; performance in processing and communications, ease of use, and tolerance of (and recovery from) faults.

Peterson, John C.↗

PARTI primitives for unstructured and block structured problems

Described here is a set of primitives (PARTI) developed to efficiently execute unstructured and block structured problems on distributed memory parallel machines. We present experimental data from a 3-D unstructured Euler solver run on the Intel Touchstone Delta to demonstrate the usefulness of our methods.

Sussman, A.↗

Optimal evaluation of array expressions on massively parallel machines

We investigate the problem of evaluating FORTRAN 90 style array expressions on massively parallel distributed-memory machines. On such machines, an elementwise operation can be performed in constant time for arrays whose corresponding elements are in the same processor. If the arrays are not aligned in this manner, the cost of aligning them is part of the cost of evaluating the expression. The choice of where to perform the operation then affects this cost. We present algorithms based on dynamic programming to solve this problem efficiently for a wide variety of interconnection schemes, including multidimensional grids and rings, hypercubes, and fat-trees. We also consider expressions containing operations that change the shape of the arrays, and show that our approach extends naturally to handle this case.

Chatterjee, Siddhartha↗

The 1987 RIACS annual report

The Research Institute for Advanced Computer Science (RIACS) was established at the NASA Ames Research Center in June of 1983. RIACS is privately operated by the Universities Space Research Association (USRA), a consortium of 64 universities with graduate programs in the aerospace sciences, under several Cooperative Agreements with NASA. RIACS's goal is to provide preeminent leadership in basic and applied computer science research as partners in support of NASA's goals and missions. In pursuit of this goal, RIACS contributes to several of the grand challenges in science and engineering facing NASA: flying an airplane inside a computer; determining the chemical properties of materials under hostile conditions in the atmospheres of earth and the planets; sending intelligent machines on unmanned space missions; creating a one-world network that makes all scientific resources, including those in space, accessible to all the world's scientists; providing intelligent computational support to all stages of the process of scientific investigation from problem formulation to results dissemination; and developing accurate global models for climatic behavior throughout the world. In working with these challenges, we seek novel architectures, and novel ways to use them, that exploit the potential of parallel and distributed computation and make possible new functions that are beyond the current reach of computing machines. The investigation includes pattern computers as well as the more familiar numeric and symbolic computers, and it includes networked systems of resources distributed around the world. We believe that successful computer science research is interdisciplinary: it is driven by (and drives) important problems in other disciplines. We believe that research should be guided by a clear long-term vision with planned milestones. And we believe that our environment must foster and exploit innovation. Our activities and accomplishments for the calendar year 1987 and our plans for 1988 are reported.

Source record↗

Unstructured grids on SIMD torus machines

Unstructured grids lead to unstructured communication on distributed memory parallel computers, a problem that has been considered difficult. Here, we consider adaptive, offline communication routing for a SIMD processor grid. Our approach is empirical. We use large data sets drawn from supercomputing applications instead of an analytic model of communication load. The chief contribution of this paper is an experimental demonstration of the effectiveness of certain routing heuristics. Our routing algorithm is adaptive, nonminimal, and is generally designed to exploit locality. We have a parallel implementation of the router, and we report on its performance.

Bjorstad, Petter E.↗

First CLIPS Conference Proceedings, volume 2

The topics of volume 2 of First CLIPS Conference are associated with following applications: quality control; intelligent data bases and networks; Space Station Freedom; Space Shuttle and satellite; user interface; artificial neural systems and fuzzy logic; parallel and distributed processing; enchancements to CLIPS; aerospace; simulation and defense; advisory systems and tutors; and intelligent control.

Source record↗

An SQL query generator for CLIPS

As expert systems become more widely used, their access to large amounts of external information becomes increasingly important. This information exists in several forms such as statistical, tabular data, knowledge gained by experts and large databases of information maintained by companies. Because many expert systems, including CLIPS, do not provide access to this external information, much of the usefulness of expert systems is left untapped. The scope of this paper is to describe a database extension for the CLIPS expert system shell. The current industry standard database language is SQL. Due to SQL standardization, large amounts of information stored on various computers, potentially at different locations, will be more easily accessible. Expert systems should be able to directly access these existing databases rather than requiring information to be re-entered into the expert system environment. The ORACLE relational database management system (RDBMS) was used to provide a database connection within the CLIPS environment. To facilitate relational database access a query generation system was developed as a CLIPS user function. The queries are entered in a CLlPS-like syntax and are passed to the query generator, which constructs and submits for execution, an SQL query to the ORACLE RDBMS. The query results are asserted as CLIPS facts. The query generator was developed primarily for use within the ICADS project (Intelligent Computer Aided Design System) currently being developed by the CAD Research Unit in the California Polytechnic State University (Cal Poly). In ICADS, there are several parallel or distributed expert systems accessing a common knowledge base of facts. Expert system has a narrow domain of interest and therefore needs only certain portions of the information. The query generator provides a common method of accessing this information and allows the expert system to specify what data is needed without specifying how to retrieve it.

Snyder, James↗

Balancing Contention and Synchronization on the Intel Paragon

The Intel Paragon is a mesh-connected distributed memory parallel computer. It uses an oblivious and deterministic message routing algorithm: this permits us to develop highly optimized schedules for frequently needed communication patterns. The complete exchange is one such pattern. Several approaches are available for carrying it out on the mesh. We study an algorithm developed by Scott. This algorithm assumes that a communication link can carry one message at a time and that a node can only transmit one message at a time. It requires global synchronization to enforce a schedule of transmissions. Unfortunately global synchronization has substantial overhead on the Paragon. At the same time the powerful interconnection mechanism of this machine permits 2 or 3 messages to share a communication link with minor overhead. It can also overlap multiple message transmission from the same node to some extent. We develop a generalization of Scott's algorithm that executes complete exchange with a prescribed contention. Schedules that incur greater contention require fewer synchronization steps. This permits us to tradeoff contention against synchronization overhead. We describe the performance of this algorithm and compare it with Scott's original algorithm as well as with a naive algorithm that does not take interconnection structure into account. The Bounded contention algorithm is always better than Scott's algorithm and outperforms the naive algorithm for all but the smallest message sizes. The naive algorithm fails to work on meshes larger than 12 x 12. These results show that due consideration of processor interconnect and machine performance parameters is necessary to obtain peak performance from the Paragon and its successor mesh machines.

Bokhari, Shahid H.↗

A High-Order Direct Solver for Helmholtz Equations with Neumann Boundary Conditions

In this study, a compact finite-difference discretization is first developed for Helmholtz equations on rectangular domains. Special treatments are then introduced for Neumann and Neumann-Dirichlet boundary conditions to achieve accuracy and separability. Finally, a Fast Fourier Transform (FFT) based technique is used to yield a fast direct solver. Analytical and experimental results show this newly proposed solver is comparable to the conventional second-order elliptic solver when accuracy is not a primary concern, and is significantly faster than that of the conventional solver if a highly accurate solution is required. In addition, this newly proposed fourth order Helmholtz solver is parallel in nature. It is readily available for parallel and distributed computers. The compact scheme introduced in this study is likely extendible for sixth-order accurate algorithms and for more general elliptic equations.

Sun, Xian-He↗

Opus: A Coordination Language for Multidisciplinary Applications

Data parallel languages, such as High Performance fortran, can be successfully applied to a wide range of numerical applications. However, many advanced scientific and engineering applications are multidisciplinary and heterogeneous in nature, and thus do not fit well into the data parallel paradigm. In this paper we present Opus, a language designed to fill this gap. The central concept of Opus is a mechanism called ShareD Abstractions (SDA). An SDA can be used as a computation server, i.e., a locus of computational activity, or as a data repository for sharing data between asynchronous tasks. SDAs can be internally data parallel, providing support for the integration of data and task parallelism as well as nested task parallelism. They can thus be used to express multidisciplinary applications in a natural and efficient way. In this paper we describe the features of the language through a series of examples and give an overview of the runtime support required to implement these concepts in parallel and distributed environments.

Chapman, Barbara↗

A Comparison of PETSC Library and HPF Implementations of an Archetypal PDE Computation

Two paradigms for distributed-memory parallel computation that free the application programmer from the details of message passing are compared for an archetypal structured scientific computation a nonlinear, structured-grid partial differential equation boundary value problem using the same algorithm on the same hardware. Both paradigms, parallel libraries represented by Argonne's PETSC, and parallel languages represented by the Portland Group's HPF, are found to be easy to use for this problem class, and both are reasonably effective in exploiting concurrency after a short learning curve. The level of involvement required by the application programmer under either paradigm includes specification of the data partitioning (corresponding to a geometrically simple decomposition of the domain of the PDE). Programming in SPAM style for the PETSC library requires writing the routines that discretize the PDE and its Jacobian, managing subdomain-to-processor mappings (affine global- to-local index mappings), and interfacing to library solver routines. Programming for HPF requires a complete sequential implementation of the same algorithm, introducing concurrency through subdomain blocking (an effort similar to the index mapping), and modest experimentation with rewriting loops to elucidate to the compiler the latent concurrency. Correctness and scalability are cross-validated on up to 32 nodes of an IBM SP2.

Hayder, M. Ehtesham↗