• Journal of the Korean Data and Information Science Society
• /
• v.26 no.5
• /
• pp.1129-1139
• /
• 2015

Inference process generates additional triples from knowledge represented in RDF triples of semantic web technology. Tens of million of triples as an initial big data and the additionally inferred triples become a knowledge base for applications such as QA(question&answer) system. The inference engine requires more computing resources to process the triples generated while inferencing. The additional computing resources supplied by underlying resource pool in cloud computing can shorten the execution time. This paper addresses an algorithm to allocate the number of computing nodes "elastically" at runtime on Hadoop, depending on the size of knowledge data fed. The model proposed in this paper is composed of the layered architecture: the top layer for applications, the middle layer for distributed parallel inference engine to process the triples, and lower layer for elastic Hadoop and server visualization. System algorithms and test data are analyzed and discussed in this paper. The model hast the benefit that rich legacy Hadoop applications can be run faster on this system without any modification.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.50-53
• /
• 2016

Image processing techniques play an increasingly important role in many aspects of our daily life. For example, it has been shown to improve agricultural productivity in a number of ways such as plant pest detecting or fruit grading. However, massive quantities of images generated in real-time through multi-devices such as remote sensors during monitoring plant growth lead to the challenges of big data. Meanwhile, most current image processing systems are designed for small-scale and local computation, and they do not scale well to handle big data problems with their large requirements for computational resources and storage. In this paper, we have proposed an IPABigData (Image Processing Algorithm BigData) platform which provides algorithms to support large-scale image processing in agriculture based on Hadoop framework. Hadoop provides a parallel computation model MapReduce and Hadoop distributed file system (HDFS) module. It can also handle parallel pipelines, which are frequently used in image processing. In our experiment, we show that our platform outperforms traditional system in a scenario of image segmentation.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.3
• /
• pp.51-58
• /
• 2007

The shared-nothing multiprocessor architecture has advantages in scalability, this architecture has been adopted in many multiprocessor database system. But, if the data are not uniformly distributed across the processors, load will be unbalanced. Therefore, the whole system performance will deteriorate. This is the data skew problem, which usually occurs in processing parallel hash join. Balancing the load before performing join will resolve this problem efficiently and the whole system performance can be improved. In this paper, we will present an algorithm using merit of very large memory to reduce disk access overhead in performing load balancing and to efficiently solve the data skew problem. Also, we will present analytical model of our new algorithm and present the result of some performance study we made comparing our algorithm with the other algorithms in handling data skew.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.32 no.1
• /
• pp.77-91
• /
• 2007

In this paper, we present a symbiotic evolutionary algorithm for multi-objective optimization. The goal in multi-objective evolutionary algorithms (MOEAs) is to find a set of well-distributed solutions close to the true Pareto optimal solutions. Most of the existing MOEAs operate one population that consists of individuals representing the entire solution to the problem. The proposed algorithm has a two-leveled structure. The structure is intended to improve the capability of searching diverse and food solutions. At the lower level there exist several populations, each of which represents a partial solution to the entire problem, and at the upper level there is one population whose individuals represent the entire solutions to the problem. The parallel search with partial solutions at the lower level and the Integrated search with entire solutions at the upper level are carried out simultaneously. The performance of the proposed algorithm is compared with those of the existing algorithms in terms of convergence and diversity. The optimization problems with continuous variables and discrete variables are used as test-bed problems. The experimental results confirm the effectiveness of the proposed algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.12
• /
• pp.5212-5230
• /
• 2016

How to determine the right number of small base stations to activate in multi-cell uplinks to match traffic from a fixed quantity of K users is an open question. This paper analyses the uplink cooperative that jointly receives base stations activation to explore this question. This paper is different from existing works only consider transmitting power as optimization objective function. The global objective function is formulated as a summation of two terms: transmitting power for data and coordinated overhead for control. Then, the joint base stations activation and beamforming problem is formulated as a mixed integer second order cone optimization. To solve this problem, we develop two polynomial-time distributed methods. Method one is a two-stage solution which activates no more than K small base stations (SBSs). Method two is a heuristic algorithm by dual decomposition to MI-SOCP that activates more SBSs to obtain multiple-antennae diversity gains. Thanks to the parallel computation for each node, our methods are more computationally efficient. The strengths and weaknesses of these two proposed two algorithms are also compared using numerical results.

• Journal of Korea Multimedia Society
• /
• v.2 no.4
• /
• pp.378-389
• /
• 1999

Parallel tasks in distributed real-time systems can be divided into several subtasks and be executed in parallel according to their real-time attributes. But, it is difficult to gain the optimal solution which is to allocate a tasks deadline into the subtasks deadline while minimizing the subtasks deadline miss. Tn this Paper, we propose the algorithm that allocates deadlines into each subtask, according to the attributes of each subtask(i.e. using communication time and execution time to periodic tasks). Also, we suggest a processor mapping algorithm that considers the communication time among the processors and the effective duplication algorithm which is allocated to the identical processor for the purpose of improving the communication time between the subtasks. We can obtain a result that reduces IPC(Inter-Processor Communication) time and uses the idle processor through applying effective real-time attributes to FUTD(Fully connected, Unbounded Task Duplication) algorithms. As a result, we can improve the average processor utilization.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.1
• /
• pp.135-145
• /
• 2010

Efficient Directed Acyclic Graph(DAG) scheduling is critical for achieving high performance in Distributed Heterogeneous computing System(DHCS). In this paper, we present a new high-performance scheduling algorithm, called the LCFT(Levelized Critical First Task) algorithm, for DHCS. The LCFT algorithm is a list-based scheduling that uses a new attribute to efficiently select tasks for scheduling in DHCS. The complexity of LCFT is $O(\upsilon+e)(p+log\;\upsilon)$. The performance of the algorithm has been observed by its application to some practical DAGs, and by comparing it with other existing scheduling algorithms such as PETS, HPS, HCPT and GCA in terms of the schedule length and SpeedUp. The comparison studies show that LCFT significantly outperforms PETS, HPS, HCPT and GCA in schedule length, SpeedUp.

• ETRI Journal
• /
• v.44 no.1
• /
• pp.62-72
• /
• 2022

In this study, we propose a multi-GPU-based 8KVR stitching system that operates in real time on both local and cloud machine environments. The proposed system first obtains multiple 4 K video inputs, decodes them, and generates a stitched 8KVR video stream in real time. The generated 8KVR video stream can be downloaded and rendered omnidirectionally in player apps on smartphones, tablets, and head-mounted displays. To speed up processing, we adopt group-of-pictures-based distributed decoding/encoding and buffering with the NV12 format, along with multi-GPU-based parallel processing. Furthermore, we develop several algorithms such as equirectangular projection-based color correction, real-time CG overlay, and object motion-based seam estimation and correction, to improve the stitching quality. From experiments in both local and cloud machine environments, we confirm the feasibility of the proposed 8KVR stitching system with stitching speed of up to 83.7 fps for six-channel and 62.7 fps for eight-channel inputs. In addition, in an 8KVR live streaming test on the 5G MEC/cloud, the proposed system achieves stable performances with 8 K@30 fps in both indoor and outdoor environments, even during motion.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.2 s.40
• /
• pp.35-42
• /
• 2006

In this paper we implement a new simulator for performance analysis of a parallel digital signal processing distributed shared memory multiprocessor systems. using Parsec The key idea of this simulator is suitable in simulation of system that uses DMA function of TMS320C6701 DSP chip and local memory which have fast access time. Also, because correction of performance parameter and reconfiguration for hardware components are easy, we can analyze performance of system in various execution environments. In the simulation, FET, 2D FET, Matrix Multiplication. and Fir Filter, which are widely used DSP algorithms. have been employed. Using our simulator, the result has been recorded according to different the number of processor, data sizes, and a change of hardware element. The performance of our simulator has been verified by comparing those recorded results.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.44 no.3
• /
• pp.74-80
• /
• 2007

Advances in electronic and computer technologies have paved the way for the proliferation of WSN(wireless sensor networks). Accordingly, necessity of anti-collusion and authentication technology is increasing on the sensor network system. Some of the algorithm developed for the anti-collision sensor network can be easily adopted to wireless sensor network platforms and in the same time they can meet the requirements for sensor networks like: simple parallel distributed computation, distributed storage, data robustness and auto-classification of sensor readings. To achieve security in wireless sensor networks, it is important to be able to establish safely channel among sensor nodes. In this paper, we proposed the USN(Ubiquitous Sensor Network) channel establishment algorithm for sensor's authentication and anti-collision. Two different data aggregation architectures will be presented, with algorithms which use wavelet filter to establish channels among sensor nodes and BIBD (Balanced Incomplete Block Design) which use anti-collision methods of the sensors. As a result, the proposed algorithm based on BIBD and wavelet filter was made for 98% collision detection rate on the ideal environment.