Danko Basch

Ontology-Based Design Pattern Recognition

This paper presents ontology-based architecture for pattern recognition in the context of static source code analysis. The proposed system has three subsystems: parser, OWL ontologies and analyser. The parser subsystem translates the input code to AST that is constructed as an XML tree. The OWL ontologies define code patterns and general programming concepts. The analyser subsystem constructs instances of the input code as ontology individuals and asks the reasoner to classify them. The experience gained in the implementation of the proposed system and some practical issues are discussed. The recognition system successfully integrates the knowledge representation field and static code analysis, resulting in greater flexibility of the recognition system.

FISCHERPLOTS: An Excel spreadsheet for computing Fischer plots of accommodation change in cyclic carbonate successions in both the time and depth domains

Fischer plots are plots of accommodation (derived by calculating cumulative departure from mean cycle thickness) versus cycle number or stratigraphic distance (proxies for time), for cyclic carbonate platforms. Although many workers have derived programs to do this, there are currently no published, easily accessible programs that utilize Excel. In this paper, we present an Excel-based spreadsheet program for Fischer plots, illustrate how the data are input, and how the resulting plots may be interpreted. The plots can be used to derive periods of increased accommodation, shown on the plots as a rising limb (which commonly matches times of more open marine, subtidal parasequence development). Times of decreased accommodation, shown on the plots as a falling limb, generally are coincident with thin, shallow, peritidal parasequences.

A high performance memory database for web application caches

This paper presents the architecture and characteristics of a memory database intended to be used as a cache engine for web applications. Primary goals of this database are speed and efficiency while running on SMP systems with several CPU cores (four and more). A secondary goal is the support for simple metadata structures associated with cached data that can aid in efficient use of the cache. Due to these goals, some data structures and algorithms normally associated with this field of computing needed to be adapted to the new environment.

Microprocessor architecture design with ATLAS

A concrete example of special purpose microprocessor development demonstrates how this simulation and analysis package facilitates the early stages of design. Simulation data enable the designer to improve performance quickly and easily with less probability of error.

Processor architecture model for fuzzy control

In this paper, the architecture of a fuzzy processor intended for high-speed control applications is described. Its main advantage is the inference speed that does not depend on the number of used inputs and rules. This is accomplished by the usage of content addressable memory as a rule set storage. The proposed architecture was simulated and the simulation results give the inference speed of about 200 KFLIPS.

A Few Notes on Variance of Frequency Estimated Applying Analytical Leakage Compensation

One of the most accurate and commonly accepted methods for spectral analysis, i.e., estimation of frequency from samples of a signal, is analytical leakage compensation, briefly called analytical solution (AS). The AS estimate is negligibly biased, but it is the whole class of solutions whose variance depends on a free parameter named K. Thus, an important step in AS algorithm is calculation of optimal Kopt, which selects the minimum variance solution from theoretically infinite set. The original paper provides complete set of needed formulae, and here a few other relations are discussed, which can be used to further reduce the variance, if properly applied according to the specified criterion.

Mark without much Sweep Algorithm for Garbage Collection

In this paper two simple improvements over traditional mark-sweep collector are proposed. The core idea is placing small objects of the same type in buckets. The buckets are organised in such way to eliminate the internal fragmentation, sweeping, and freeing inside them. The measured improvement of garbage collection time over traditional mark-sweep is 19%. Another proposed improvement is more general and is applicable to other garbage collection algorithms as well. It uses heuristics to control the heap growth. The regularities in behaviour of objects of particular types are used to determine whether the collection should be performed or avoided in favour of immediate heap expansion. The heap expansion algorithm reduces garbage collection time over traditional mark-sweep for 49% while keeping the heap size approximately the same.

A Note on Optimality of Analytical Leakage Compensation at Boundary Frequencies

The interpolated fast Fourier transform (IFFT) was one of the first methods for the highly accurate estimation of sine wave parameters, and its first successful descendant was analytical leakage compensation [which is commonly called analytical solution (AS)]. The AS estimate of frequency is a whole class of solutions whose variance depends on a free parameter K. Thus, to extract the minimum variance solution from a theoretically infinite set, we have to find the optimal value K opt. This paper clarifies the mathematical background of AS and proposes two new solutions for K opt, which reduce the variance of AS estimates of low and high frequencies close to an integer. All inferences are justified by simulations, which confirm the validity of theoretical considerations.