• Journal of KIISE:Information Networking
• /
• v.32 no.6
• /
• pp.705-715
• /
• 2005

A peer-to-peer web caching has been studied recently as it can reduce the traffic converged on the server side and can support the traditional web caching model. Although the peer-to-peer web caching has the merit of having additional cache space from the local caches of peers without additional infrastructure, several constraints such as dynamic participation and local caching strategy caused by the autonomy of peers in peer-to-peer networks nay limit the performance of the peer-to-peer web caching. To overcome these limitations, we propose an efficient directory-based peer-to-peer web caching system under dynamic participation of peers. In the proposed caching system, we present new peer selection and replica management schemes by introducing the concept of the object lifetime in P2P networks. We evaluate the effectiveness of the proposed system through trace-driven simulations with a web log dataset. Simulation results show that the proposed system has higher accuracy and fewer redirection failures than the conventional directory-based P2P web caching system in feasible peer-to-peer networks.

• Journal of Korean Society on Water Environment
• /
• v.24 no.5
• /
• pp.603-610
• /
• 2008

A mathematical model was developed to understand how the presence of plants affects vertical profiles of electron acceptors, their reduced species, and trace metals in the wetland sediments. The model accounted for biodegradation of organic matter utilizing sequential electron acceptors and subsequent chemical reactions using stoichiometric relationship. These biogeochemical reactions were affected by the combined effects of oxygen release and evapotranspiration driven by wetland plants. The measured data showed that $SO_4{^{2-}}$ concentrations increased at the beginning of the growing season and then gradually decreased. Based on the measured data, it was hypothesized that the limitation of the solid phase sulfide in direct contact with the roots may result in the gradual decrease of $SO_4{^{2-}}$ concentrations. With the dynamic formulation for the limitation of the solid phase sulfide, model simulated time variable sulfate profiles using published model parameters. Oxygen release from roots produced divalent metal species (i.e. $Cd^{2+}$) as well as oxidized sulfur species (i.e. $SO_4{^{2-}}$) in the sediment pore water. Evapotranspiration-induced advection increased flux of divalent metal species from the overlying water column into the rhizosphere. The increased divalent metal species were converted to the metal sulfide with sufficient FeS around the rhizosphere, which contributed to the decrease of bioavailability and toxicity of divalent metal activity in the pore water. Since the divalent metal activity is a good predictor of the metal bioavailability, this model with a proper simulation of solid phase sulfide plays an essential role to predict the dynamics of trace metals in the wetland sediments.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.6
• /
• pp.1266-1272
• /
• 1997

Due to the price and performanceof uniprocessor workstations and off-the shelf networking, network of workstations(NOW) ae now a cost-effective parallel processing platform tht is competitive with supercomputers. Meanwhile, current network fiile system protocols rely heavily on a central server to coordinate file activity among client workstations. This central server can become a bottleneck that limits scalibility for environments with large numbers of clients. In this paper, we propsoe a highly reliable and effective software RAID file system on the network of workstation environment. We present results form a trace-driven simulation study that shows that the designed software RAID file system is more effective in the aspect of elapsed time when compared with client/server file systems.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2003.05a
• /
• pp.1179-1185
• /
• 2003

시뮬레이션 상의 입력모델에 대한 기존의 연구는 과거의 자료를 바탕으로 선형의 모수적인 (parametric) 모형을 개발하는데 초점을 두고 있다. 그러나 이 경우에는 입력이 매우 복잡한 형태를 가지면 모수적인 모형을 잦는 것이 불가능해지므로 비모수적인(non-parametric) 접근방법이 절실한 실정이다 예로 인터넷 트래픽 모델의 시뮬레이션 수행시 입력으로 제공되는 단위 시간당 요구되는 웹 페이지의 수 같은 경우 데이터들 간데 종속관계가 매우 심하고 복잡하여 모수적 모형을 세우는데 어려움이 있다. 이러한 시스템들을 시뮬레이션 방법으로 분석 하고자 할 때, 기존의 trace-driven 시뮬레이션 방법이나 모수적 모형을 찾아 다수의 사실적인 시뮬레이션 입력 자료를 확보하는 것은 현실적으로 어려움이 있다. 따라서. 비모수적인 방법으로 다수의 사실적인 시뮬레이션 입력 자료를 생성하는 것이 필요하다. 이러한 비모수적인 방법에 대한 평가기준 설정은 시뮬레이션 상의 입력 모델에 대한 타당성을 제시한다는 점에서 또한 매우 중요하다. 본 논문에서는 붓트스트 랩의 방법중의 하나인 임계값 붓트스트랩을 이용하여 시뮬레이션 입력 자료 생성 방법을 개발하였고 Turing test를 통해 붓스트랩으로 생성산 입력 시나리오를 검증하였다.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.13 no.5
• /
• pp.245-252
• /
• 2018

Non-volatile RAM devices have been increasingly viewed as an alternative of DRAM main memory system. However some technologies including phase-change memory (PCM) are still suffering from relatively poor write performance as well as limited endurance. In this paper, we introduce a proactive last-level cache management to efficiently hide a low write performance of non-volatile main memory systems. The proposed method significantly reduces the cache miss penalty by proactively evicting the part of cachelines when the non-volatile main memory system is in idle state. Our trace-driven simulation demonstrates 24% performance enhancement, compared with a conventional LRU cache management, on the average.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.1009-1012
• /
• 1999

An analytical performance model that can predict the performance of a superscalar processor employing multiple branch prediction is introduced. The model is based on the conditional independence probability and the basic block size of instructions, with the degree of multiple branch prediction, the fetch rate, and the window size of a superscalar architecture. Trace driven simulation is performed for the subset of SPEC integer benchmarks, and the measured IPCs are compared with the results derived from the model. As the result, our analytic model could predict the performance of the superscalar processor using multiple branch prediction within 6.6 percent on the average.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.12
• /
• pp.1704-1709
• /
• 2014

In order to increase the performance of multi-core system processor architectures, the multi-thread branch predictor which speculatively fetches and allocates threads to each core should be highly accurate. In this paper, the perceptron based multi-thread branch predictor is proposed for the multi-core processor architectures. Using SPEC 2000 benchmarks as input, the trace-driven simulation has been performed for the 2 to 16-core architectures employing perceptron multi-thread branch predictor extensively. Its performance is compared with the architecture which utilizes the two-level adaptive multi-thread branch predictor.

• Proceeding of EDISON Challenge
• /
• 2016.03a
• /
• pp.473-476
• /
• 2016

In this study, important design factors in Jansen leg mechanism by which two legs can be driven by only one input like a motor are considered through method of transmitting motion in three-bar linkage and Grashof law in four-bar linkage. In preliminary design, by using EDISON m-sketch and its simulation which can observe trace of feet, two identical four-bar linkages are initially designed and two three-bar linkages are added to four-bar linkages sequentially. By analyzing GL(Ground Length) and GAC(Ground Angle Coefficient), the adequacy of the preliminary design was estimated. Final design of walking apparatus is implemented using CAD software, Assembly2 of EDISON Designer. Finally, proposals to improve software used in this study are suggested.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.10
• /
• pp.1563-1566
• /
• 2022

Video streaming applications such as Netflix and Youtube are widely used in our daily life. A DASH based streaming client exploits adaptive bit rate (ABR) method to choose the most appropriate video source representation that the network can support. In this paper we propose a novel energy-aware ABR scheme that adds the ability to monitor energy efficiency in addition to the linear quadratic regulator algorithm we previously introduced. Our trace-driven simulation studies show that our proposed scheme mitigates and shortens re-buffering, resulting in energy savings of mobile devices while preserving the similar QoE compared to the state-of-the-art ABR algorithms.

• Journal of the Korea Computer Industry Society
• /
• v.5 no.1
• /
• pp.57-64
• /
• 2004

The speed gap between processors and disks is a serious problem. So, I/O sub-system limits the performance of computer system. To overcome the speed gap, caches have been used in computer system. By using cache, the access times to disk blocks can be reduced and the performance of computer system can be improved. In this paper, we proposed an efficient cache management algorithm for computer system which have buffer cache and disk cache. The proposed algorithm can minimize the duplicated blocks between buffet cache and disk cache. We evaluate the proposed algorithm by trace-driven simulation. The simulation results show that the proposed algorithm can reduce the mean access time to disk blocks.