• Journal of the Korea Society of Computer and Information
• /
• v.19 no.2
• /
• pp.39-48
• /
• 2014

In this paper, we propose an efficient dynamic workload balancing strategy which improves the performance of high-performance computing system. The key idea of this dynamic workload balancing strategy is to minimize execution time of each job and to maximize the system throughput by effectively using system resource such as CPU, memory. Also, this strategy dynamically allocates job by considering demanded memory size of executing job and workload status of each node. If an overload node occurs due to allocated job, the proposed scheme migrates job, executing in overload nodes, to another free nodes and reduces the waiting time and execution time of job by balancing workload of each node. Through simulation, we show that the proposed dynamic workload balancing strategy based on CPU, memory improves the performance of high-performance computing system compared to previous strategies.

• Journal of KIISE:Databases
• /
• v.35 no.1
• /
• pp.67-83
• /
• 2008

In order to realize ubiquitous computing, it is essential to efficiently use the resources and the computing power of mobile devices. Among others, memory efficiency, energy efficiency, and processing efficiency are required in executing the softwares embedded in mobile devices. In this paper, query processing over XML data in a mobile device where resources are limited is addressed. In a device with limited amount of memory, the techniques of XML. stream query processing need to be employed to process queries over a large volume of XML data Recently, a technique Galled XFrag was proposed whereby XML data is fragmented with the hole-filler model and streamed in fragments for processing. With XFrag, query processing is possible in the mobile device with limited memory without reconstructing the XML data out of its fragment stream. With the hole-filler model, however, memory efficiency is not high because the additional information on holes and fillers needs to be stored. In this paper, we propose a new technique called XFLab whereby XML data is fragmented with the XML labeling scheme which is for representing the structural relationship in XML data, and streamed in fragments for processing. Through implementation and experiments, XML showed that our XFLab outperformed XFrag both in memory usage and processing time.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.5
• /
• pp.537-541
• /
• 2008

Log-Structured File System(LFS) is a disk based file system that is optimized for improving the write performance. LFS gathers dirty data in memory as long as possible, and flushes all dirty data sequentially at once. In a real system, however, maintaining dirty data in memory should be flushed into a disk to meet file system consistency issues even if more memory is still available. This synchronizations increase the cleaner overhead of LFS and make LFS to write down more metadata into a disk. In this paper, by adapting Non-volatile RAM(NV-RAM) we modifies LFS and virtual memory subsystem to guarantee that LFS could gather enough dirty data in the memory and reduce small disk writes. By doing so, we improves the performance of LFS by around 2.5 times than the original LFS.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.7
• /
• pp.708-712
• /
• 2008

Existing dynamic memory allocation schemes for general purpose operating system can not directly apply to the wireless sensor networks (WSNs). Because these schemes did not consider features of WSNs, they consume a lot of energy and waste the memory space caused by fragmentation. In this paper, we found features of WSNs applications and made the model which adapts these issues. Through this research, we suggest the slab allocator that reduces the execution time and the memory management space. Also, we evaluate the performance of our scheme by comparing to one of the previous systems.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.40 no.11
• /
• pp.972-978
• /
• 2012

This paper discusses the parallelization of multifrontal solution method, widely used for finite element structural analyses, for a shared memory architecture. Multifrontal method is easier than other linear solution methods because the solution procedure implies that unknowns can be eliminated simultaneously. Two innovative ideas are introduced to achieve optimal solver performance on a shared memory computer. Those are pairing two frontal matrices and splitting the frontal matrix in order to reduce the temporal memory space required by independent computing tasks. Performance comparisons between original algorithm and proposed one prove that proposed method is more computationally efficient on current multicore machines.

• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.6
• /
• pp.367-372
• /
• 2013

Cloud computing technology has been developing at an increasing expansion rate. Today most of firms are using this technology, making improving the quality of service one of the most important issues. To achieve this, the system must operate efficiently with less idle time and without deteriorating the customer satisfaction. This paper focuses on enhancing the efficiency of a conventional Genetic Algorithm (GA) for task scheduling in cloud computing using Fuzzy Logic (FL). This study collected a group of task schedules and assessed the quality of each task schedule with the user expectation. The work iterates the best scheduling order genetic operations to make the optimal task schedule. General GA takes considerable time to find the correct scheduling order when all the fitness function parameters are the same. GA is an intuitive approach for solving problems because it covers all possible aspects of the problem. When this approach is combined with fuzzy logic (FL), it behaves like a human brain as a problem solver from an existing database (Memory). The present scheme compares GA with and without FL. Using FL, the proposed system at a 100, 400 and 1000 sample size*5 gave 70%, 57% and 47% better improvement in the task time compared to GA.

• Korean Journal of Computational Design and Engineering
• /
• v.16 no.3
• /
• pp.236-245
• /
• 2011

Many researchers have recently studied multi-level formulation strategies to solve the MDO problems and they basically distributed the coupling compatibilities across all disciplines, while single-level formulations concentrate all the controls at the system-level. In addition, approximation techniques became remedies for computationally expensive analyses and simulations. This paper studies comparisons of the MDO methods with respect to computing performance considering both conventional sequential and modem distributed/parallel processing environments. The comparisons show Individual Disciplinary Feasible (IDF) formulation is the most efficient for sequential processing and IDF with approximation (IDFa) is the most efficient for parallel processing. Results incorporating to popular design examples show this finding. The author suggests design engineers should firstly choose IDF formulation to solve MDO problems because of its simplicity of implementation and not-bad performance. A single drawback of IDF is requiring more memory for local design variables and coupling variables. Adding cheap memories can save engineers valuable time and effort for complicated multi-level formulations and let them free out of no solution headache of Multi-Disciplinary Analysis (MDA) of the Multi-Disciplinary Feasible (MDF) formulation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.11
• /
• pp.3619-3637
• /
• 2022

Big data analytics offers endless opportunities for operational enhancement by extracting valuable insights from complex voluminous data. Hadoop is a comprehensive technological suite which offers solutions for the large scale storage and computing needs of Big data. The performance of Hadoop is closely tied with its configuration settings which depends on the cluster capacity and the application profile. Since Hadoop has over 190 configuration parameters, tuning them to gain optimal application performance is a daunting challenge. Our approach is to extract a subset of impactful parameters from which the performance enhancing sub-optimal configuration is then narrowed down. This paper presents a statistical model to analyze the significance of the effect of Hadoop parameters on a variety of performance metrics. Our model decomposes the total observed performance variation and ascribes them to the main parameters, their interaction effects and noise factors. The method clearly segregates impactful parameters from the rest. The configuration setting determined by our methodology has reduced the Job completion time by 22%, resource utilization in terms of memory and CPU by 15% and 12% respectively, the number of killed Maps by 50% and Disk spillage by 23%. The proposed technique can be leveraged to ease the configuration tuning task of any Hadoop cluster despite the differences in the underlying infrastructure and the application running on it.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.689-692
• /
• 2021

This paper is a prior study to improve the efficiency of offloading based on mobile agents to optimize allocation of computing resources and reduce latency that support user proximity of application services in a Fog/Edge Computing (FEC) environment. We propose an algorithm that effectively reduces the execution time and the amount of memory required when extracting optimal moving patterns from the vast set of spatio-temporal movement history data of moving objects. The proposed algorithm can be useful for the distribution and deployment of computing resources for computation offloading in future FEC environments through frequency-based optimal path extraction.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.2 s.344
• /
• pp.51-59
• /
• 2006

Von Neumann architecture suffers from the interface between the central processing unit and the memory, which is called 'Von Neumann bottleneck' In this paper, we propose a scalable general-purpose associative processor (AP) based on content-addressable memory (CAM) which solves this problem and is suitable for the search-oriented applications. We propose an efficient instruction set and a structural scalability to extend for larger applications. We define twelve instructions and provide some reduced instructions to speed up which execute two instructions in a single instruction cycle. The proposed AP performs in a bit-serial, word-parallel fashion and can be considered as a 32-bit general-purpose parallel processor with a massively parallel SIMD structure. We design and simulate a maximum/minumum search greater-than/less-than search, and parallel addition to verify the proposed architecture. The algorithms are executed in a constant time O(k) regardless of the number of input data.