• Journal of Service Research and Studies
• /
• v.4 no.2
• /
• pp.11-19
• /
• 2014

This study simulates cases with 20, 30, 40, 50, 60 and 100 students and 6 values as the parameter of mu. Using the simulation, an analysis was made of how the number of students that couldn't concentrate during class increased over the duration of time. When normalizing the difference in student number when the value of mu is constant, there was a decrease in percentage over time. It also showed a change in percentage when the value of time is fixed while the value of mu changes. Based on the result values, by estimating the mu value with the class atmosphere to predict the outcome and improve the lecture.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.9
• /
• pp.585-591
• /
• 2006

Generalized Processor Sharing(GPS) model provides instantaneous fair services to currently backlogged sessions. Since this fair service distributes server capacity to backlogged sessions in proportion to their weights, the fairness is only valid between the sessions serviced at the same time. From the long time view, however, this fair service provides different server capacity with one session or another, even if these sessions have the same weights. This paper proposes a cumulative fair service(CFS) model to provide fair server capacity to all sessions in the long time view. This model provides fair service in session Viewpoint because it distributes server capacity in proportion to the weights of sessions. The model and an algorithm referencing that model are analyzed for their properties and performances. Performance evaluations verify that the proposed algorithm provides proportional service capacity to sessions in the long time view.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.7A
• /
• pp.987-995
• /
• 1999

The handoff is a distinctive characteristic of the mobile networks. In the CDMA systems, if base stations support multiple Frequency Assignment (FA), they provide both soft handoff and hard handoff. Under the CDMA environments, the soft handoff guarantees the favorable service quality and the continuity of call connection without interruption, and increases the service capacity of the base stations. This paper proposes call connection control schemes with handoff queue for supporting efficient handoff processing. The proposed schemes are divided into two categories: single handoff queue scheme and multiple handoff queue scheme. We analyze the performance of the proposed call connection control schemes using numerical analysis. From the analysis results, we can say that it is more desirable to avoid hard handoff as long as handoff queues are used. When a single handoff queue used, adaptive scheme that properly mixes avoidable and avoidable hard handoff method under the given traffic condition is more desirable. In case that multiple handoff queues are used, the suitable trade-off needs to be developed between handoff blocking probability and hard handoff probability to guarantee a given blocking probability threshold.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.44 no.5
• /
• pp.7-15
• /
• 2007

In wireless communication networks, single-radio single-channel architecture degrades throughput and end-to-end delay due to half-duplex transmission of wireless node and route intra interference. Also, In contention-based MAC (Medium Access Control) architecture, channel contention reduces throughput and packet collision enlarges end-to-end delay. In this paper, we use multi-radio multi-channel architecture which will make wireless node to operate in full duplex mode, and exclude route intra interference. Based on this architecture, we propose a new link metric, ccf which reflects the characteristics of a contention-based wireless link, and propose a routing path metric MCCR considering channel switching delay and route intra interference. MCCR is compared with MCR by simulation, the performance of a route established by MCCR outperforms the performance of a route by MCR in terms of throughput and end-to-end delay.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.7
• /
• pp.847-862
• /
• 1999

본 논문에서는 유한버퍼의(finite-buffered) 동기화된(synchronous) 큐잉모델(queueing model)을 이용하여 명령어들간의 병렬성, 분기명령의 빈도수, 분기예측(branch prediction)의 정확도, 캐쉬미스 등의 파라미터들을 고려하여 프로세서의 명령어 실행율을 예측하며 캐쉬의 성능과 파이프라인 성능간의 관계를 분석할 수 있는 새로운 해석적 모델을 제안하였다. 해석적 모델은 모델의 타당성을 검증하기 위해서 시뮬레이션을 수행하여 얻은 결과와 비교하였다. 해석적 모델과 시뮬레이션을 비교한 결과 대부분 10% 오차 내에서 일치하였다. 본 연구를 통하여 얻은 해석적 모델을 사용하면 시뮬레이션에서는 드러나지 않는 성능제약의 원인에 대한 명확한 규명이 가능하기 때문에 성능향상을 위한 설계자료를 얻을 수 있으며, 시스템 성능 밸런스를 위한 캐쉬와 비순차이슈 파이프라인 성능간의 관계에 대한 정확한 분석이 가능하다.Abstract This research presents a novel analytic model to predict the instruction execution rate of superscalar processors using the queuing model with finite-buffer size and synchronous operation mode. The proposed model is also able to analyze the performance relationship between cache and pipeline. The proposed model takes into account various kinds of architectural parameters such as instruction-level parallelism, branch probability, the accuracy of branch prediction, cache miss, and etc.. To prove the correctness of the model, we performed extensive simulations and compared the results with the analytic model. Simulation results showed that the proposed model can estimate the average execution rate accurately within 10% error compared to simulation results. The proposed model can explain the causes of performance bottleneck which cannot be uncovered by the simulation method only. The model is also able to show the effect of the cache miss on the performance of out-of-order issue superscalar processors, which can provide an valuable information in designing a balanced system.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.4
• /
• pp.171-180
• /
• 2004

CDP(handoff Call Dropping Probability) and CBP(new Call Blocking Probability) have been used as two important call level QoS parameters in cellular mobile communications. But, many methods to reduce CDP without considering CBP have been studied, and hand-off call priority scheme has been introduced. But the use of hand-off call priority scheme increases CBP and decreases channel utilization rate depending on the number of reserved channel for priority. In this paper, we propose a CAC(Call Admission Control) algorithm with overflow and preemption to solve the problem caused by considering CDP and CBP in calculation of the number of channel reserved. The problem is the increase of CDP as the traffic load increases. In our CAC algorithm, hand-off call is permitted to use(overflow) unreserved and unused channel if there is no reserved and unused channel, and new call is permitted to use(preemption) the channel overflowed by hand-off call if there is no unreserved and unused channel. This mechanism of calculation of the number of reserved channel and CAC algorithm is expected to increase channel utilization rate, and can be applied to media-based QoS provision in cellular mobile communications.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.12
• /
• pp.73-82
• /
• 2008

There are several approaches for reducing the ROB(Reorder Buffer) complexity in processors. The one technique that makes the simplest ROB ports relies on a distributed implementation that spreads the centralized ROB structure across the functional units(FUs). Each distributed buffers are decided on the size of them by workload of the functional units. The performance of the processor depends on the size of distributed ROB. However, most of previous works have depended on the simulation results to decide the optimsize of distributed ROB. In this Paper, we use an analytical model based on the M/M/1 Queuing theory to determine the optimum size of each distributed ROB. Our schemes are evaluated by using the simulation performed by the CPU2000 benchmarks. We are able to choose the optimum size of distributed ROB showing the 99.2% performance compared with existing superscalar processors. We can save 82% hardware resources in ports and reduce more than 30% of delay when ROB and distributed ROB proposed in this paper are designed by HDL.