• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.4B
• /
• pp.255-269
• /
• 2006

The differentiated services(DiffServ) architecture provides packet level service differentiation through the simple and predefined Per-Hop Behaviors(PHBs). The Assured Forwarding(AF) PHB proposed as the assured services uses the RED-in/out(RIO) approach to ensusre the expected capacity specified by the service profile. However, the AF PHB fails to give good QoS and fairness to the TCP flows. This is because OUT(out- of-profile) packet droppings at the RIO buffer are unfair and sporadic during only network congestion while the TCP's congestion control algorithm works with a different round trip time(RTT). In this paper, we propose an Adaptive Regulating Drop(ARD) marker, as a novel dropping strategy at the ingressive edge router, to improve TCP fairness in assured services without a decrease in the link utilization. To drop packets pertinently, the ARD marker adaptively changes a Temporary Permitted Rate(TPR) for aggregate TCP flows. To reduce the excessive use of greedy TCP flows by notifying droppings of their IN packets constantly to them without a decrease in the link utilization, according to the TPR, the ARD marker performs random early fair remarking and dropping of their excessive IN packets at the aggregate flow level. Thus, the throughput of a TCP flow no more depends on only the sporadic and unfair OUT packet droppings at the RIO buffer in the core router. Then, the ARD marker regulates the packet transmission rate of each TCP flow to the contract rate by increasing TCP fairness, without a decrease in the link utilization.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.2A
• /
• pp.98-106
• /
• 2005

We propose a new wireless fair queueing algorithm, WFQ-R (Wireless Fair Queueing with Retransmission), which is well matched with the LLR (Link Level Retransmission) algorithm and does not require channel prediction. In the WFQ-R algorithm, the share consumed by retransmission is regarded as a debt of the retransmitted flow to the other flows. Thus, the WFQ-R algorithm achieves wireless fairness with the LLR algorithm by penalizing flows that use wireless resources without permission under the MAC layer. Through simulations, we showed that our WFQ-R algorithm maintains fairness adaptively and maximizes system throughput. Furthermore, our WFQ-R algorithm is able to achieve flow separation and compensation.

• Journal of information and communication convergence engineering
• /
• v.9 no.4
• /
• pp.411-419
• /
• 2011

In this paper, we propose an Adaptive Regulating Drop (ARD) marker, as a novel dropping strategy at the ingressive edge router, to improve TCP fairness in assured services (ASs) without a decrease in the link utilization. To drop packets pertinently, the ARD marker adaptively changes a Temporary Permitted Rate (TPR) for aggregate TCP flows. The TPR is set larger than the current input IN packet rate of aggregate TCP flows while inversely proportional to the measured input OUT packet rate. To reduce the excessive use of greedy TCP flows by notifying droppings of their IN packets constantly to them without a decrease in the link utilization, the ARD marker performs random early fair remarking of their excessive IN packets to OUT packets at the aggregate flow level according to the TPR. In addition, an aggregate dropper is combined to drop some excessive IN packets fairly and constantly according to the TPR. Thus, the throughput of a TCP flow no more depends on only the sporadic and unfair OUT packet droppings at the RIO buffer in the core router. Then, the ARD marker regulates the packet transmission rate of each TCP flow to the contract rate by increasing TCP fairness, without a decrease in the link utilization.

• Journal of Korea Multimedia Society
• /
• v.18 no.12
• /
• pp.1475-1482
• /
• 2015

In network routers, buffers are used to resolve congestion and reduce packet loss rate whenever congestion occurs at bottleneck link. Most of the existing methods to manage such buffers focus only on queue-length-based control as one loop which have some issues of low link utilization and system stability. In this paper, we propose a novel framework which exploits two-loop control method, e.g. queue-length and congestion window size, combined with optimal margin method to facilitate parameter choices. Simulation results in ns-2 demonstrate that bottleneck link performance can be improved with higher link utilization (85%) and shorter queue length (22%) than the current deployed scheme in commercial routers (RED and DropTail).

• Journal of the Korean Operations Research and Management Science Society
• /
• v.23 no.3
• /
• pp.49-62
• /
• 1998

In the ring loading problem, traffic demands are given for each pair of nodes in an undirected ring network with n nodes and a flow is routed in either of the two directions, clockwise and counter-clockwise. The load of a link is the sum of the flows routed through the link and the objective of the Problem is to minimize the maximum load on the ring. In the ring loading problem with integer demand splitting, each demand can be split between the two directions and the flow routed in each direction is restricted to integers. Recently, Vachani et al. ［INFORMS J. Computing 8 (1996) 235-242］ have developed an Ο(n$^3$) algorithm for solving this integer version of the ring loading problem and independently, Schrijver et al. ［to appear in SIAM J. Disc. Math.］ have presented an algorithm which solves the problem with ｛0,1｝ demands in Ο（n$^2$｜K｜ ) time where K denotes the index set of the origin-desㅇtination pairs of nodes having flow demands. In this paper, we develop an algorithm which solves the problem in Ο(n ｜K｜) time.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.12 no.4
• /
• pp.193-204
• /
• 2017

Various 3D-stacked DRAMs have been proposed to overcome the memory wall problem. Hybrid Memory Cube (HMC) is a true 3D-stacked DRAM with stacked DRAM layers on top of a logic layer. The logic die is mainly used to implement a memory controller for HMC, and it is connected through a high speed serial link called SerDes with a host that is either a processor or another HMC. In HMC, the serial link is crucial for both performance and power consumption. Therefore, it is important that the link is configured properly so that the required performance should be satisfied while the power consumption is minimized. In this paper, we propose a HMC system model included the high speed serial link to estimate performance accurately. Since the link modeling strictly follows the link flow control mechanism defined in the HMC spec, the actual HMC performance can be estimated accurately with respect to each link configuration. Various simulations are conducted in order to deduce the correlation between the HMC performance and the link configuration with regard to memory utilization. It is confirmed that there is a strong correlation between the achievable maximum performance of HMC and the link configuration in terms of both bandwidth and latency. Therefore, it is possible to find the best link configuration when the required HMC performance is known in advance, and finding the best configuration will lead to significant power saving while the performance requirement is satisfied.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2006.11a
• /
• pp.599-617
• /
• 2006

A network model and a Genetic Algorithm(GA) is proposed to solve the simultaneous estimation of the trip distribution and traffic assignment from traffic counts in the congested networks in a logit-based Stochastic User Equilibrium (SUE). The model is formulated as a problem of minimizing the non-linear objective functions with the linear constraints. In the model, the flow-conservation constraints of the network are utilized to restrict the solution space and to force the link flows meet the traffic counts. The objective of the model is to minimize the discrepancies between the link flows satisfying the constraints of flow-conservation, trip production from origin, trip attraction to destination and traffic counts at observed links and the link flows estimated through the traffic assignment using the path flow estimator in the legit-based SUE. In the proposed GA, a chromosome is defined as a vector representing a set of Origin-Destination Matrix (ODM), link flows and travel-cost coefficient. Each chromosome is evaluated from the corresponding discrepancy, and the population of the chromosome is evolved by the concurrent simplex crossover and random mutation. To maintain the feasibility of solutions, a bounded vector shipment is applied during the crossover and mutation.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.21 no.6
• /
• pp.544-551
• /
• 2008

We have studied the dielectric characteristics of low-k interlayer dielectric materials fabricated by PECVD for various precursor's flow rates. BTMSM precursor was introduced with the flow rates from 42 sccm to 60 sccm by 2 sccm step in the constant flow rate of 60 sccm $O_2$. The absorption intensities of Si-O-$CH_x$ bonding group and Si-$CH_x$ bonding group changed synchronously for the variation of precursor flow rate, but the intensity of Si-O-Si(C) responded asynchronously with the $CH_x$ combined bonds. The heat treatment reduced the FTIR absorption intensity of Si-O-$CH_x$ bonding group and Si-$CH_x$ bonding group but increased the intensity of Si-O-Si(C). The nanopore and free space formed by the increasement of caged link mode and cross link mode of Si-O-Si(C) group implied the origin of low-k SiOCH films.

• Proceedings of the Korean Statistical Society Conference
• /
• 2002.11a
• /
• pp.119-124
• /
• 2002

We consider a multi-state flow network consisted of undirected links and focus on how to find efficiently the union of minimal paths transmitting a required flow when minimal paths are known.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.11
• /
• pp.1726-1734
• /
• 1993

This paper propose a survivable communication network design process using node degree that augments the usual traffic flow and cost analyses with previously ignored topological survivability and computing time considerations. At first, decide a initial topology, and then measure a throughput of network. If the throughput is smaller than the required traffic, add edge to the optimum place by using minimum node degree and link distance. Otherwise, drop useless edge by using maximum node degree, link distance and link utilization. This process is repeated until throughput equals to the required traffics. This Process designs a survivable communication network with the minimized cost and computing time and usual traffic flow. The design proceses that minimized computing time are freely select initial topology and easily design a large network. And these results of algorithm are compared with the Kris and Pramod's in order to analyses the perfmance of the designed network.