• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.6
• /
• pp.318-330
• /
• 2002

This paper discusses the design and implementation of a video server which uses tertiary device, such as magneto-optical jukebox or tape library, as a source of media archiving. In order to handle the tertiary device in the framework of disk-based stream service model, a sophisticated streaming model is required for server and it should consider the device-specific performance characteristics of tertiary storage. We have carefully designed the streaming mechanism for server whose key functionalities include stream scheduling, disk caching and admission control. The proposed system has been implemented on Windows 2000 platform which is equipped with a magneto-optical jukebox. The design principles of the server are validated with the experiments and the performance characteristics is analyzed.

• The Journal of the Korea Contents Association
• /
• v.14 no.5
• /
• pp.8-17
• /
• 2014

Petascale systems(so called supercomputers) have been mainly used for supporting communication-intensive and tightly-coupled parallel computations based on message passing interfaces such as MPI(HPC: High-Performance Computing). On the other hand, computing paradigms such as High-Throughput Computing(HTC) mainly target compute-intensive (relatively low I/O requirements) applications consisting of many loosely-coupled tasks(there is no communication needed between them). In Korea, recently emerging applications from various scientific fields such as pharmaceutical domain, high-energy physics, and nuclear physics require a very large amount of computing power that cannot be supported by a single type of computing resources. In this paper, we present our HTCaaS(High-Throughput Computing as a Service) which can leverage national distributed computing resources in Korea to support these challenging HTC applications and describe the details of our system architecture, job execution scenario and case studies of various scientific applications.

• Bulletin of the Korean Space Science Society
• /
• 2003.10a
• /
• pp.46-46
• /
• 2003

Since its launching on 21 December 1999, the KOrea Multi-Purpose SATellite-Ⅰ (KOMPSAT-Ⅰ) has been successfully operated by the Mission Control Element (MCE), which was developed by the Electronics and Telecommunications Research Institute (ETRI). Most of the major functions of the MCE have been successfully demonstrated and verified during the three years of the mission life of the satellite. The Mission Analysis and Planning Subsystem (MAPS), which is one of the four subsystems in the MCE, played a key role in the Launch and Early Orbit Phase (LEOP) operations as well as the on-orbit mission operations. This paper presents the operational performances of the various functions in MAPS. We show the performance and analysis of orbit determinations using ground-based tracking data and GPS navigation solutions. We present four instances of the orbit maneuvers that guided the spacecraft from injection orbit into the nominal on-orbit. We include the ground-based attitude determination using telemetry data and the attitude maneuvers for imaging mission. The event prediction, mission scheduling, and command planning functions in MAPS subsequently generate the spacecraft mission operations and command plan. The fuel accounting and the realtime ground track display also support the spacecraft mission operations. We also present the orbital evolutions during the three years of the mission life of the KOMPSAT-Ⅰ.

• Journal of KIISE
• /
• v.43 no.6
• /
• pp.613-620
• /
• 2016

This paper presents a MOHA(Many-Task Computing on Hadoop) framework which aims to effectively apply the Many-Task Computing(MTC) technologies originally developed for high-performance processing of many tasks, to the existing Big Data processing platform Hadoop. We present basic concepts, motivation, preliminary results of PoC based on distributed message queue, and future research directions of MOHA. MTC applications may have relatively low I/O requirements per task. However, a very large number of tasks should be efficiently processed with potentially heavy inter-communications based on files. Therefore, MTC applications can show another pattern of data-intensive workloads compared to existing Hadoop applications, typically based on relatively large data block sizes. Through an effective convergence of MTC and Big Data technologies, we can introduce a new MOHA framework which can support the large-scale scientific applications along with the Hadoop ecosystem, which is evolving into a multi-application platform.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.54 no.2
• /
• pp.147-156
• /
• 2012

가뭄강도와 지속기간을 고려하여 관개구역을 몇 개의 관개블록으로 나누어 순차적으로 제한하여 급수함으로써 관개용수를 절약하고 관개기간을 연장하여 가뭄을 극복하는 윤환관개모형 (RISS)을 제안한 바 있다. 이 RISS는 작물생육기별 관개저수지 저수율별 급수구역과 제한 급수율을 제시함으로써 가뭄기간에 저수지 운영자가 언제, 얼마나 제한하여 급수할 것인지 제시해주는 모형으로 전북 남원군 동화댐의 물 관리시스템 자동화 프로그램에 2004년 설치하여 운용되도록 하였다. 이 연구에서는 2004년부터 2009년까지 동화 댐 저수위 관측기록과 급수실적을 분석하여 RISS의 절수효과를 평가하고자 하였다. 그러나 실제로는 동화댐의 물 관리시스템 자동화 프로그램이 계획대로 운용되지 않고 있었다. 따라서 이 RISS의 절수효과와 효용성을 입증하여 계획대로 자동화시스템을 유용하게 활용하도록 하기 위해서 급수실적 및 저수위 관측기록과 RISS로 모의 발생한 저수위 결과를 비교 평가하였다. 모의발생 평가결과 가뭄이 극심했던 2007년에는 3.8백만 $m^3$을 절수하여 관개기간을 23일 연장하고 2009년에는 3.3백만 $m^3$을 절수하여 관개기간을 13일 연장하여 가뭄이 극심한 해에는 RISS 모형의 효용성이 크다는 것을 알 수 있었다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.11B
• /
• pp.1016-1022
• /
• 2003

The EPON has been actively studied as one of the access networks for the economic configuration of FTTH. The EPON must support the dynamic bandwidth allocation to the subscribers in order to support the QoS due to its base on Ethernet technology EFM SG, which is actively working for the standardization of EPON, also recently decided to select DBA. Therefore in this paper, we designed a ONU buffer scheduling algorithm(AIWFQ) and a scheme of DBA(Class-based FCFS) for the OLT suitable for embodying MPCP of the EPON. In this paper, we proposed methods that the EPON system can make use of by measuring end to end process delay time and the buffer size in order to implement the algorithm by using the OPNET.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37A no.11
• /
• pp.979-985
• /
• 2012

The design of an efficient scheduler is a key design factor in IEEE 802.16m systems, in order to support services with various QoS smoothly. Although conventional studies of schedules have been suggested, those had problems that are not able to satisfy the delay condition and make the through-put declined, because they only focused on a specific action scenario rather than reflecting practical action scenarios which have real-time and non-real-time traffics variously. In this paper, an efficient uplink scheduling algorithm is proposed for IEEE 802.16m system by introducing the concepts of Virtual Time(VT) and Virtual Finish Time(VFT) based priority determination, and separate buffers for each QoS class in the mobile terminal. Simulation results showed that the proposed scheme had satisfied the delay requirement of real-time services even with improved throughput performance compared to conventional methods.

• Journal of Information Processing Systems
• /
• v.6 no.3
• /
• pp.413-434
• /
• 2010

Wireless sensor networks are usually characterized by dense deployment of energy constrained nodes. Due to the usage of a large number of sensor nodes in uncontrolled hostile or harsh environments, node failure is a common event in these systems. Another common reason for node failure is the exhaustion of their energy resources and node inactivation. Such failures can have adverse effects on the quality of the real-time services in Wireless Sensor Networks (WSNs). To avoid such degradations, it is necessary that the failures be recovered in a proper manner to sustain network operation. In this paper we present a dynamic Energy efficient Real-Time Job Allocation (ERTJA) algorithm for handling node failures in a cluster of sensor nodes with the consideration of communication energy and time overheads besides the nodes' characteristics. ERTJA relies on the computation power of cluster members for handling a node failure. It also tries to minimize the energy consumption of the cluster by minimum activation of the sleeping nodes. The resulting system can then guarantee the Quality of Service (QoS) of the cluster application. Further, when the number of sleeping nodes is limited, the proposed algorithm uses the idle times of the active nodes to engage a graceful QoS degradation in the cluster. Simulation results show significant performance improvements of ERTJA in terms of the energy conservation and the probability of meeting deadlines compared with the other studied algorithms.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.10 no.2
• /
• pp.66-76
• /
• 2011

The Sightseeing ITS is defined as a system that gives tourists and inhabitants the information concerned with road network conditions, parking, sightseeing spots, and travel time in the relevant area. When focusing on tourists' excursion behaviors in sightseeing areas, however, it is sometimes found that they feel much annoyed with insufficient information about the area. While they also have a variety of needs to get information and make a decision about their excursions, it becomes one of important subjects to understand the actual circumstances of their information usage and to identify the causal relationship between their excursion patterns and the obtained information. This paper aims to investigate basic issues toward development of Sightseeing ITS and to modify an analytical model of tourist's excursion behavior considering information usage at Fuji five lakes Area, which is a representative sightseeing place in Japan. The relationship was analyzed using a structural equation model: information usage at the scheduling stage has an impact on tourists' individual cognition of the sightseeing area, tourists' individual cognition affects information usage at the excursion stage, and information usage at the excursion stage affects behavior during excursions. It is found that tourists' decision making behaviors on their excursion tend to depend on the information usage. This implies essential results to construct effective and customer oriented ITS.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.9
• /
• pp.1551-1557
• /
• 2006

Since subscribers, who pay the same communication fees, have a strong desire to take similar services with the other users, it is an important issue to provide fairness among those subscribers wherever they are located. Therefore, in this paper we propose a QoS packet scheduler that can provide fairness among wireless LAN terminals and evaluate its performance using computer simulation. The key idea of this scheduler is to reduce the CW value of the wireless LAN terminal that has failed in sending its packets due to channel transmission error in order to offer the wireless MM terminal a higher transmission priority which is the idea of service compensation. We evaluate its performance using NS-2 network simulator, compare its numerical results to those of IEEE 802.11e without this scheme, and conclude that this scheme can reduce throughput difference between similar wireless LAN terminals.