فهرست مقالات parallelization

• دسترسی آزاد مقاله
  • صفحه چکیده
  • متن کامل
  
  1 - Parallelization of Rich Models for Steganalysis of Digital Images using a CUDA-based Approach
  Mahmoud Kazemi Meysam Mirzaee Reza Isfahani
  There are several different methods to make an efficient strategy for steganalysis of digital images. A very powerful method in this area is rich model consisting of a large number of diverse sub-models in both spatial and transform domain that should be utilized. Howev چکیده کامل

  There are several different methods to make an efficient strategy for steganalysis of digital images. A very powerful method in this area is rich model consisting of a large number of diverse sub-models in both spatial and transform domain that should be utilized. However, the extraction of a various types of features from an image is so time consuming in some steps, especially for training phase with a large number of high resolution images that consist of two steps: train and test. Multithread programming is a near solution to decreasing the required time but it’s limited and it ‘snot so scalable too. In this paper, we present a CUDA based approach for data-parallelization and optimization of sub-model extraction process. Also, construction of the rich model is analyzed in detailed, presenting more efficient solution. Further, some optimization techniques are employed to reduce the total number of GPU memory accesses. Compared to single-thread and multi-threaded CPU processing, 10x-12x and 3x-4x speedups are achieved with implementing our CUDA-based parallel program on GT 540M and it can be scaled with several CUDA cards to achieve better speedups. پرونده مقاله
• دسترسی آزاد مقاله
  • صفحه چکیده
  • متن کامل
  
  2 - Comparison of information transfer delay in standard Apriori algorithm and improved Apriori algorithm
  Hooman Bavarsad Salehpour Seyed Hamid Seyed Javadi Parvaneh Asghari Mohammad Ebrahim Shiri Ahmad Abadi
  One of the most famous algorithms in the field of focused exploration of data mining correlation rules is the Apriori algorithm and its many developed versions. But what can be raised as a major challenge in this field is the proper application of this algorithm in the چکیده کامل

  One of the most famous algorithms in the field of focused exploration of data mining correlation rules is the Apriori algorithm and its many developed versions. But what can be raised as a major challenge in this field is the proper application of this algorithm in the distributed environments of today's world. In this research, a parallelization-based approach is proposed to improve the performance of the Apriori algorithm in the process of exploring recurring patterns on network topologies. The proposed approach includes two major features: (1) combining the node centrality criterion and the Apriori algorithm to identify frequent patterns, (2) using the mapping/reduction method in order to create parallel processing and achieve optimal values in the shortest time. Also, this approach pursues three main goals: reducing the temporal and spatial complexity of the Apriori algorithm, improving the process of extracting dependency rules and identifying recurring patterns, comparing the performance of the proposed approach on different network topologies in order to determine the advantages and disadvantages of each topology. To prove the superiority of the proposed method, a comparison has been made between our approach and the basic Apriori algorithm. The evaluation results of the methods prove that the proposed approach provides an acceptable performance in terms of execution time criteria compared to other methods. پرونده مقاله