The Community for Technology Leaders
Reconfigurable Computing and FPGAs, International Conference on (2011)
Cancun, Quintana Roo Mexico
Nov. 30, 2011 to Dec. 2, 2011
ISBN: 978-0-7695-4551-6
pp: 74-79
Data-mining over software can reveal similar patterns on software code. This can give important insights for the design of hardware cores, especially considering the benefits of the merge of software kernels and their implementation as a single hardware core. However, software codes have characteristics that make inadequate the direct use of typical data mining tools, mainly related to their large number of samples and the imprecise definition of code features for mining. Those characteristics affect negatively the performance of the most known data mining methods. To solve this problem, we propose in this paper the use of three techniques: the Normalized Compression Distance, the Neighbor Joining, and the Fast Newman algorithm. We combine these three techniques and propose a new approach for data mining of code repositories (DAMICORE). DAMICORE works with different types of code representations. Experiments reveal DAMICORE can indicate important software similarities at source code level. Specifically, merging soft-ware kernels identified by DAMICORE results in FPGA cores with size smaller than the overall hardware size needed when implementing a core for each kernel.
Data Mining, Normalized Compression Distance, Neighbor Joining, Fast Newman, Hardware Cores, FPGA, Data-path Merging

A. Sanches, A. C. Delbem and J. M. Cardoso, "Identifying Merge-Beneficial Software Kernels for Hardware Implementation," 2011 International Conference on Reconfigurable Computing and FPGAs (ReConFig 2011)(RECONFIG), Cancun, 2011, pp. 74-79.
179 ms
(Ver 3.3 (11022016))