International Conference Papers

Permanent URI for this collectionhttp://dl.cerist.dz/handle/CERIST/4

Browse

Author: search.filters.author.Bendjoudi, AhcèneSubject: search.filters.subject.GPU Computing

Search Results

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Multi and Many-core Parallel B&B approaches for the Blocking Job Shop Scheduling Problem
    (2016-07) Dabah, Adel; Bendjoudi, Ahcène; Ait Zai, Abdelhakim; El Baz, Didier; Nouali-Taboudjemat, Nadia
    In this paper, we propose three approaches to accelerate the B&B execution time using Multi and Many-core systems to solve the NP-hard Blocking Job Shop Scheduling problem (BJSS). The first approach is based on Master/Worker paradigm where the workers independently explore the branches sent by the master. The second approach is a node-based parallelization that does not change the design of the B&B algorithm, except that the bounding process is faster since it is calculated in parallel using several threads organized in one GPU block. The third approach is a Multi-Core CPU/GPU hybridization that benefits from the power of both the CPU-cores and the GPU at the same time. This hybridization is based on concurrent kernels execution provided by Nvidia Multi process Service (MPS) i.e. each host process (Master or Worker) launches his own kernel to accelerate the bounding process on GPU. The obtained results using Taillard instances confirm the efficiency of our proposals. The first two approaches are respectively three and eighteen times faster compared to the sequential version. The results of the hybrid approach show a relative speedup over ninety times as compared to the sequential approach and therefore prove the advantage of using both the CPU-cores and the GPU at the same time.
  • Thumbnail Image
    Item
    Data reordering for minimizing threads divergence in GPU-based evaluating association rules
    (2015-06) Djenouri, Youcef; Bendjoudi, Ahcène; Mehdi, Malika; Habbas, Zineb; Nouali-Taboudjemat, Nadia
    This last decade, the success of Graphics Processor Units (GPUs) has led researchers to launch a lot of works on solving large complex problems by using these cheap and powerful architecture. Association Rules Mining (ARM) is one of these hard problems requiring a lot of computational resources. Due to the exponential increase of data bases size, existing algorithms for ARM problem become more and more inefficient.Thus, research has been focusing on parallelizing these algorithms. Recently, GPUs are starting to be used to this task. However, their major drawback is the threads divergence problem. To deal with this issue, we propose in this paper an intelligent strategy called transactions-based Reordering ”TR” allowing an efficient evaluation of association rules on GPU by minimizing threads divergence. This strategy is based on data base re-organization. To validate our proposition, theoretical and experimental studies have been carried out using well-known synthetic datasets. The results are very promising in terms of minimizing the number of threads divergence.
  • Thumbnail Image
    Item
    Parallel Rules Mining Using GPUs and Bees Behaviors
    (2014-08) Djenouri, Youcef; Bendjoudi, Ahcène; Mehdi, Malika; Nouali-Taboudjemat, Nadia; Habbas, Zineb
    This paper addresses the problem of association rules mining with large data sets using bees behaviors. The bees swarm optimization method have been successfully applied on small and medium data size. Nevertheless, when dealing Webdocs benchmark (the largest benchmark on the web), it is bluntly blocked after more than 15 days. Additionally, Graphic processor Units are massively threaded providing highly intensive computing and very usable by the optimization research community. The parallelization of such method on GPU architecture can be deal large data sets as the case of WebDocs in real time. In this paper, the evaluation process of the solutions is parallelized. Experimental results reveal that the suggested method outperforms the sequential version at the order of ×100 in most data sets, furthermore, the WebDocs benchmark is handled with less than ten hours.