Biplab Kumer Sarker

A GA BASED MULTIPLE TASK ALLOCATION CONSIDERING LOAD

A Distributed Computing System (DCS) comprising networked heterogeneous processors requires ecient tasks to processor allocation to achieve minimum turnaround time and highest possible throughput. Task allocation in DCS remains an important and relevant problem attracting the attention of researchers in the discipline. A good number of task allocation algorithms have been proposed in the literature [3{9]. This algorithm considered allocation of the modules of a single task to various processing nodes and aim to minimize the turnaround time of the given task. But they did not consider execution of modules belonging to various dierent tasks (i.e. multiple tasks). In this work we have considered the number of modules that can be accepted by individual processing nodes along with their memory capacities and arrival of multiple disjoint tasks to the DCS from time to time. In this paper, a method based on genetic algorithm is developed which is memory ecient and give an optimal solution of the problem. The given simulation results also show signicant achievement in this regard.

Scheduling in Distributed Computing Systems

No wonder you activities are, reading will be always needed. It is not only to fulfil the duties that you need to finish in deadline time. Reading will encourage your mind and thoughts. Of course, reading will greatly develop your experiences about everything. Reading scheduling in distributed computing systems is also a way as one of the collective books that gives many advantages. The advantages are not only for you, but for the other peoples with those meaningful benefits.

Cluster-based multiple task allocation in distributed computing system

Summary form only given. Most of the task allocation models & algorithms in distributed computing system (DCS) require a priori knowledge of its execution time on the processing nodes. Since the task assignment is not known in advance, this time is quite difficult to estimate. We propose a cluster-based dynamic allocation scheme, in a distributed computing system, which eliminate this time requirement. Further, as opposed to a single task allocation, generally proposed in most of the models, we consider multiple tasks. A fuzzy function is used for both the module clustering and processor clustering. Dynamic invocation of clustering and assignment is considered. Experimental results show the efficacy of the proposed model.

Parallel algorithms for mining association rules in time series data

A tremendous growing interest in finding dependency among patterns has been developing in the domain of time series data mining. It is quite effective to find how current and past values in the streams of data are related to the future. However, these kind of data sets with high dimensionality are enormous in size results in possibly large number of mined dependencies. This strongly motivates the need of efficient parallel algorithms. In this paper, we propose two parallel algorithms to discover dependency from the large amount of time series data. We introduce the method of extracting sequence of symbols from the time series data by using segmentation and clustering processes. To reduce the search space and speed up the process we investigate the technique to group the time series data. The experimental results conducted on a shared memory multiprocessors system justifies the inevitability of using parallel techniques for mining huge amount of data in the time series domain.

Towards a weighted-tree similarity algorithm for RNA secondary structure comparison

A tree similarity algorithm for RNA (ribonucleic acid) secondary structure comparison is presented. The elements (nucleotides and nucleotide-pairs) of an RNA secondary structure are represented as normalized node-weighted trees. We show that our weighted tree representations of RNA secondary structures are informative and useful. Based on this unique representation for RNA secondary structure, we propose a weighted-tree similarity algorithm for computing the similarity between RNA secondary structures. The algorithm is justified by computing similarities among several well-known RNA secondary structures. For a given RNA secondary structure, the proposed algorithm provides a ranked list of RNA structures in a database according to their similarity values with the query RNA. Hence, our algorithm is helpful in predicting the functions and the class of a newly discovered RNA

Allocation aspects in Distributed Computing Systems

Distributed Computing System (DCS) is the natural candidate for the future computing system and in recent past it has received much attention in the computing community. Before the system is fully available for use, certain Important issues are to be resolved. An essential phase in Operating System (OS) of the DCS is task allocation, which is an NP-Hard problem. Several task allocation algorithms for Distributed Computing Systems have been reported in the literature[1-5]. These algorithms are based on certain assumptions. We have analysed the implications of the assumptions of these algorithms. Another important issue, in allocation, is the even distribution of the load onto the processing nodes of the DCS (Load balancing). This work aims at identification of a proper task allocation problem model for Distributed Computing System. The work is expected to set guidelines for the researchers of the discipline and get an enlightened consensus over these issues.

Comparative study of two GA based task allocation models in distributed computing system

Genetic algorithm has emerged as a successful tool for optimization problems. Previously, we proposed a task allocation model to maximize the reliability of a distributed computing system (DCS) using a genetic algorithm. Our objective is to use the same simple genetic algorithm to minimize the turnaround time of the task given to the DCS for execution and then to compare the resultant allocation with the allocation of maximal reliability as obtained by Vidyarthi et al. (2001). Comparisons of both the algorithms are made by illustrated examples and appropriate comments.

A distributed algorithm for outlier detection in a large database

This paper proposes a distributed algorithm to detect outliers for large and distributed datasets. The algorithm employs the basis of distance-based outliers based on the distance of a point to its kth nearest neighbor. It declares the top n points in the ranking to be outliers. To the best of our knowledge, this is the first proposal of a distributed algorithm for outlier detection for shared-nothing multiple processor computing environments. It has four phases. First, in each processing node, the algorithm partitions the input data set into disjoint subsets, then it prunes entire partitions as soon as it is determined that they cannot contain outliers. Then it applies a global filtering technique to collect the partitions as global candidates from local candidate partitions in each processing node. Further, it introduces a load balancing algorithm to balance the number of local candidate partitions. Finally, it identifies outliers from each processing node.

Mining association rules from multi-stream time series data on multiprocessor systems

Mining association rules from multi-stream data has received a lot of attention to the data mining community. It is quite effective and useful to discover such rules. However, it is a very time consuming and expensive task to mine the rules from these kinds of time ordered real valued continuous data sets with high dimensionality when they are enormous in size. This strongly motivates the need of efficient parallel processing techniques and algorithms. In this paper, we use parallel processing to discover dependency from the large amount of time series multi-stream data. We apply two parallel programming techniques (OpenMP and MPI) to implement this. The experimental results conducted in multiprocessor systems show the effectiveness of MPI over OpenMp.

Load balanced allocation of multiple tasks in a distributed computing system

A Distributed Computing Systems (DCS) calls for proper partitioning of tasks into modules and allocating them to various nodes so as to enable parallel execution of their modules by individual different processors of the system. A number of algorithms have been proposed for allocation of tasks in a Distributed Computing System. Most of the models proposed in literature consider modules of a single task for static allocation, for the purpose of allocation onto a DCS. Moreover, they did not consider the architectural capability of the processing nodes and the way of connectivity among them. This work considers allocation of disjoint multiple tasks with their corresponding modules and proposes a parallel algorithm for a realistic situation wherein multiple disjoint tasks with their modules compete for execution on an arbitrarily connected DCS based on well-known A* technique. The proposed algorithm also considers a load balanced allocation for the purpose. The paper justifies the effectiveness of the algorithm with the experimental results by comparing with previously reported works.