Font Size: a A A

The Graphing Of Mathematics Image Base On Heterogeneous Computing With OpenCL

Posted on:2014-09-06Degree:MasterType:Thesis
Country:ChinaCandidate:Z H ZhouFull Text:PDF
GTID:2268330425955637Subject:Subject teaching
Abstract/Summary:PDF Full Text Request
At recent years, with the increasing popularity of computers and rapid development ofinformation technology, the modern information technology used in mathematics teaching isincreasingly strengthened. The research of information technology used in mathematics teaching isdeepening. Due to its lower price, lower power consumption, higher floating-point computationalcapacity and wider bandwidth, Graphic Process Unit(GPU) is developed rapidly, and has achievedmuch success in the fields of dense computation, physical simulation and parallel computing etc.Since the algorithm of graphing for mathematical formulae with two free variables isinefficiency, A improved algorithm of graphing for mathematical formulae with two free variablesused OpenCL technology is proposed in the paper based on the character of parallelism of itsgraphing. On the basis of a GPU being a coprocessor of CPU, the algorism is master/slave pattern.The master (CPU) initializes the shared buffer or GPU memory area, the algorithms of graphing formathematical formulae with two free variables are organized as kernel mode(the kernel of GPU), andeach graphic process units runs the same kernel instance to calculate the data. The master get thefunction image data from the shared buffer or GPU memory area after finished the graphing. In thisway, the algorithm of graphing for mathematical formulae with two free variables is performedparallelized by the graphic process unit with strong floating-point computational capacity.The experimental result shows that the speed of improved algorithm of graphing formathematical formulae with two free variables is accelerated greatly than that of traditional serialalgorithm under the condition that the precision is not changed. At the same time, the problem ofinefficiency most of the computer algebra algorithm is solved. For difference algorithms of computeralgebra system, parallel improvement appropriate can accelerate the algorithm, so OpenCL is generalpurpose and valuable practically for computer algebra system. At the same time, it also improves the efficiency of math lesson preparation and teaching.
Keywords/Search Tags:mathematics teaching, information technology, computer algebra systems, heterogeneous computing, OpenCL
PDF Full Text Request
Related items