Learn More
iii ABSTRACT Since processors spend most of the time in executing loop nests because of the dependencies, minimizing dependencies across loops is vital for compiler optimization. This paper explores two methods, sums of data dependencies and dominant data dependencies for eliminating dependencies in multi-dimensional loops. The first method eliminates(More)
In this study, we apply data mining tools to generate interesting patterns for predicting box office performance of movies using data collected from multiple social media and web sources including Twitter, YouTube and the IMDb movie database. The prediction is based on decision factors derived from a historical movie database, followers count from Twitter,(More)
Real-time tasks for command and control systems are too large or too complex for one processor to handle. Simply adding more CPUs does not result in a linear increase in performance. Current comparative analysis of parallel algorithms does not accurately reflect the increased cost of scheduling when more processors are added. A case is made that associative(More)
When dealing with randomly located or clustered data, interpolation error will vary as the distance to the nearest sample or cluster of samples. The current predominant methods for interpolating non-uniform data are not guaranteed to handle this variability of error well. The non-uniformity of the error surface can easily lead to gross misinterpretations of(More)