• Journal of information and communication convergence engineering
• /
• v.9 no.4
• /
• pp.471-474
• /
• 2011

In parallel job scheduling, each job can be executed simultaneously on multiple machines at a time. Thus in the input instance, a job $J_i$ requires the number $m_i$ of machines on which it shall be processed. The algorithm should determine not only the execution order of jobs but also the machines on which the jobs are executed. In this paper, when the jobs have deadlines, the problem is to maximize the total work of jobs which is completed by their deadlines. The problem is known to be strongly NP-hard [5] and we investigate the approximation algorithms for the problem. We consider a model in which the algorithm can have more machines than the adversary. With this advantage, the problem is how good solution the algorithm can produce against the optimal algorithm.

• Journal of IKEEE
• /
• v.2 no.2 s.3
• /
• pp.247-254
• /
• 1998

This paper presents a performance study of two LU decomposition algorithms on two massively parallel SIMD machines: the 16K processor MasPar MP-1 and the 4K processor MasPar MP-2. The paper presents experimental results and an analysis of the algorithms to explain the results. While the blocked and the nonblocked algorithms for LU decomposition have been studied individually by others, we compare the two algorithms and identify the tradeoffs between them. Our analysis of the blocked algorithm shows how the block size affects the interprocessor communication cost and the memory read/write overhead. The analysis in this paper is useful to determine an optimum block size for the blocked algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.6C
• /
• pp.438-443
• /
• 2008

The efficient method to implement the Max-Log-MAP algorithm is proposed by modifying the conventional algorithm. It is called a parallel soft output Viterbi algorithm (SOVA) and the rigorous proof is given for the equivalence between the Max-Log-MAP algorithm and the parallel SOVA. The parallel SOVA is compared with the conventional algorithms and we show that it is an efficient algorithm implementing the modified SOVA in parallel.

• Structural Engineering and Mechanics
• /
• v.14 no.6
• /
• pp.625-647
• /
• 2002

Parallel execution of computational mechanics codes requires efficient mesh-partitioning techniques. These mesh-partitioning techniques divide the mesh into specified number of submeshes of approximately the same size and at the same time, minimise the interface nodes of the submeshes. This paper describes a new mesh partitioning technique, employing Genetic Algorithms. The proposed algorithm operates on the deduced graph (dual or nodal graph) of the given finite element mesh rather than directly on the mesh itself. The algorithm works by first constructing a coarse graph approximation using an automatic graph coarsening method. The coarse graph is partitioned and the results are interpolated onto the original graph to initialise an optimisation of the graph partition problem. In practice, hierarchy of (usually more than two) graphs are used to obtain the final graph partition. The proposed partitioning algorithm is applied to graphs derived from unstructured finite element meshes describing practical engineering problems and also several example graphs related to finite element meshes given in the literature. The test results indicate that the proposed GA based graph partitioning algorithm generates high quality partitions and are superior to spectral and multilevel graph partitioning algorithms.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.4A
• /
• pp.271-281
• /
• 2005

This paper for establishing the reliability of the TPC command is introduced, where the soft symbol of the TPC command itself is directed used as a reliability indicator. In addition to the new reliability estimation, the concept of parallel use of TPC algorithms is presented. The results show that the soft symbol reliability estimation decrease the $P_{tx}$ levels with 0.3 dB, thus providing a useful capacity gain. The parallel use of 2 to 4 algorithms is also shown to decrease the sensitivity of the algorithms to the algorithm thresholds used, and thus increase the feasibility of the algorithms in a real world networks.

• The Journal of Information Technology and Database
• /
• v.2 no.2
• /
• pp.87-109
• /
• 1995

The index file has been used a access database records effectively. The join operation in a relational database system requires a large execution time, especially in the case of handling large size tables. If the indexes are available on the joining attributes for both relations involved in the join and the join selectivity is relatively small, we can improve the execution time of the join operation. In this paper. we investigate the performance trade-offs of parallel index-based join algorithms where different indexing schemes are used. We also present a comparison of our index-based parallel join algorithms with the hash-based parallel join algorithm.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1997.10a
• /
• pp.166-169
• /
• 1997

For parallel processing, the compiler partitions a loaded program into a set of tasks and makes a schedule for the tasks that will minimize parallel processing time for the loaded program. Building an optimal schedule for a given set of partitioned tasks of a program has known to be NP-complete. In this paper we introduce a GA(Genetic Algorithm)-based scheduling method in which a chromosome consists of two parts of a string which decide the number and order of tasks on each processor. An additional computation is used for feasibility constraint in the chromosome. By granularity theory, a partitioned program is categorized into coarse-grain or fine-grain types. There exist good heuristic algorithms for coarse-grain type partitioning. We suggested another GA adaptive to the coarse-grain type partitioning. The infeasibility of chromosome is overcome by the encoding and operators. The number of processors are decided while the GA find the minimum parallel processing time.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.23 no.3
• /
• pp.24-29
• /
• 2009

In case of the optimization problems that have many design variables, the conventional genetic algorithms(GA) fall into a trap of local minima with high probability. This problem is called the premature convergence problem. To overcome it, the parallel genetic algorithms which adopt the migration mechanism have been suggested. But it is hard to determine the several parameters such as the migration size and the migration interval for the parallel GAs. Therefore, we propose a new method to determine the migration interval automatically in this paper. To verify its validity, it is applied to some traditional mathematical optimization problems and is compared with the conventional parallel GA. It is also applied to the optimal design of the brushless DC motor for an electric wheel chair which is a real world problem and has five design variables.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.9
• /
• pp.4349-4371
• /
• 2019

Moldable parallel tasks are widely used in different areas, such as weather forecast, biocomputing, mechanical calculation, and so on. Considering the deadline and the speedup, scheduling moldable parallel tasks becomes a difficulty. Past work majorly focuses on the LA (List Algorithms) or OMA (Optimizing the Middle Algorithms). Different from prior work, our work normalizes execution time and makes all tasks have the same scope in normalized execution time: [0,1], and then according to the normalized execution time, a method is used to search for the reference execution time without considering the deadline of tasks. According to the reference execution time, we get an initial scheduling result based on AFCFS (Adaptive First Comes First Served) policy. Finally, a heuristic approach is used to improve the performance of the initial scheduling result. We call our method HSRET (a Heuristic Scheduling method based on Reference Execution Time). Comparisons to other methods show that HSRET has good performance in AWT (Average Waiting Time), AET (Average Execution Time), and PUT (Percentages of Unfinished Tasks).

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.469-470
• /
• 2018

Object tracking is a common task in computer vision, an essential part of various vision-based applications. After several years of development, object tracking in video is still a challenging problem because of various visual properties of objects and surrounding environment. Particle filter is a well-known technique among common approaches, has been proven its effectiveness in dealing with difficulties in object tracking. However, particle filter is a high-complexity algorithms, which is an severe disadvantage because object tracking algorithms are required to run in real time. In this research, we utilize parallel programming to accelerate particle filter-based object tracking algorithms. Experimental results showed that our approach reduced the execution time significantly.