Physical Sciences and Mathematics Commons^™

Building Segment Trees In Parallel, Peter Su, Scot Drysdale

Computer Science Technical Reports

The segment tree is a simple and important data structure in computational geometry [7,11]. We present an experimental study of parallel algorithms for building segment trees. We analyze the algorithms in the context of both the PRAM (Parallel Random Access Machine) and hypercube architectures. In addition, we present performance data for implementations developed on the Connection Machine. We compare two different parallel alforitms, and we also compare our parallel algorithms to a good sequential algorithm for doing the same job. In this way, we evaluate the overall efficiency of our parallel methods. Our performance results illustrates the problems involved in …

Using Minimal And Maximal Fault Tolerance For The Assessment Of Fault-Tolerant Algorithms, Martina Schollmeyer, Bruce M. Mcmillin

Computer Science Technical Reports

No abstract provided.

Relaxing Synchronization In Distributed Simulated Annealing, Chul-Eui Hong, Bruce M. Mcmillin

Computer Science Technical Reports

Simulated annealing is an attractive, but expensive, heuristic for approximating the solution to combinatorial optimization problems. Since simulated annealing is a general purpose method, it can be applied to the broad range of NP-complete problems such as the traveling salesman problem, graph theory, and cell placement with a careful control of the cooling schedule.

Attempts to parallelize simulated annealing, particularly on distributed memory multicomputers, are hampered by the algorithm’s requirement of a globally consistent system state. In a multicomputer, maintaining the global state S involves explicit message traffic and is a critical performance bottleneck. One way to mitigate this bottleneck …

A Process-Learning Pid Controller Algorithm Utilizing An On-Line Iterative Improvement Technique, Ryan Rosandich, Ralph W. Wilkerson

Computer Science Technical Reports

Three iterative improvement algorithms are presented for the determination of process controller gains. An offline algorithm is developed and tested as a basis for comparison, and a simple on-line algorithm is developed as an incremental step toward the final algorithm, proportional on-line iterative improvement. The algorithms are based on an Artificial Neural Network learning method, and this method is compared with other control optimization techniques. The performance of each of the algorithms was experimentally evaluated in numerous realistically simulated process control situations consisting of flow, level, and temperature control loops with various values of dead-time and process noise. The experimental …

Algorithms For Closest Point Problems: Practice And Theory, Peter Su

Computer Science Technical Reports

This paper describes and evaluates know sequential algorithms for constructing planar Voronoi diagrams and Delaunay triangulations. In addition, it describes a new incremental algorithm which is simple to understand and implement, but whose performance is competitive with all known methods. The experiments in this paper are more than just simple benchmarks, they evaluate the expected performance of the algorithms in a precise and machine independent fashion. Thus, the paper also illustrates how to use experimental tools to both understand the behaviour of different algorithms and to guide the algorithm design process.

Effect Of The X² Test On Construction Of Id3 Decision Trees, Mayank Thakore, Daniel C. St. Clair

Computer Science Technical Reports

Inductive machine learning algorithms are knowledge-based learning algorithms which take training instances as input and produce knowledge as output. One popular induction algorithm is Quinlan's ID3 [1986]. This algorithm produces knowledge in the form of a decision tree. Each path in the tree can be interpreted as a rule with the leaves representing rule conclusions. Selected attributes which describe the training instances form the interior nodes of the tree.

The ID3 algorithm is extremely sensitive to noisy training data. In an effort to reduce the effects of noise on tree construction, Quinlan used the X² test to identify noisy …

Formal Generation Of Executable Assertions For Application-Oriented Fault Tolerance, Hanan Lutfiyya, Martina Schollmeyer, Bruce M. Mcmillin

Computer Science Technical Reports

Executable assertions embedded into a distributed computing system can provide run-time assurance by ensuring that the program state, in the actual run-time environment, is consistent with the logical stage specified in the assertions; if not, then an error has occurred and a reliable communication of this diagnostic information is provided to the system such that reconfiguration and recovery can take place. Application- oriented fault tolerance is a method that provides fault detection using executable assertions based on the natural constraints of the application.

This paper focuses on giving application-oriented fault tolerance a theoretical foundation by providing a mathematical model for …

A Visualization System For Correctness Proofs Of Graph Algorithms, Peter A. Gloor, Donald B. Johnson, Fillia Makedon, Panagiotis Metaxas

Computer Science Technical Reports

In this paper we describe a system for visualizing correctness proofs of graph algorithms. The system has been demonstrated for a greedy algorithm. Prim's algorithm for finding a minimum spanning tree of an undirected, weighted graph. We believe that our system is particularly appropriate for greedy algorithms, though much of what we discuss can guide visualization of proofs in other contexts. While an example is not a proof, our system provides concrete examples to illustrate the operation of the algorithm. These examples can be referred to by the user interactively and alternatively with the visualization of the proof where the …

Multiprocessor File System Interfaces, David Kotz

Computer Science Technical Reports

Increasingly, file systems for multiprocessors are designed with parallel access to multiple disks, to keep I/O from becoming a serious bottleneck for parallel applications. Although file system software can transparently provide high-performance access to parallel disks, a new file system interface is needed to facilitate parallel access to a file from a parallel application. We describe the difficulties faced when using the conventional (Unix-like) interface in parallel applications, and then outline ways to extend the conventional interface to provide convenient access to the file for parallel programs, while retaining the traditional interface for programs that have no need for explicitly …

How To Encrypt /Usr/Dict/Words In About A Second, Peter Su, Matt Bishop

Computer Science Technical Reports

We present an implementation of the Data Encryption Standard on the Connection Machine architecture. The DES encryption algorithm is ideally suited to the Connection Machine because it consists of bit serial operations, and thousands of encryptions can be done in parallel, independently of one another. Thus, our code encrypts passwords about ten times faster than the fastest competition that we know about. In addition, the nature of the Connection Machine's architecture is such that some of the optimizations that make DES run much faster on conventional architectures have no effect on the performance of the Connection Machine. Our comparison of …

On The De Bruijn Torus Problem, Glenn Hurlbert, Garth Isaak

Computer Science Technical Reports

A (kn;n)k-de Bruijn Cycle is a cyclic k-ary sequence with the property that every k-ary n-tuple appears exactly once contiguously on the cycle. A (kr, ks; m, n)k-de Bruijn Torus is a k-ary krXks toroidal array with the property that every k-ary m x n matrix appears exactly once contiguously on the torus. As is the case with de Bruijn cycles, the 2-dimensional version has many interesting applications, from coding and communications to pseudo-random arrays, spectral imaging, and robot self-location. J.C. Cock proved the existence of such tori for all m, n, and k, and Chung, Diaconis, and Graham asked …

Concurrent Local Search For Fast Proximity Algorithms On Parallel And Vector Architectures, Peter Su

Computer Science Technical Reports

This paper presents a fast algorithm for solving the all-nearest-neighbors problem. The algorithm uses a data parallel style of programming which can be efficiently utilized on a variety of parallel and vector architectures [4,21,26]. I have implemented the algorithm in C on one such architecture, the Cray Y-MP. On one Cray CPU, the implementation is about 19 times faster than a fast sequential algorithm running on a Sparc workstation. The main idea in the algorithm is to divide the plane up into a fixed grid of cells, or buckets. When the points are well distributed, the algorithm processes each query …