• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.10
• /
• pp.529-538
• /
• 2007

Fair Queueing algorithms based on Generalized Processor Sharing (GPS) not only guarantee sessions with service rate and delay, but also provide sessions with instantaneous fair sharing. This fair sharing distributes server capacity to currently backlogged sessions in proportion to their weights without regard to the amount of service that the sessions received in the past. From a long-term perspective, the instantaneous fair sharing leads to a different quality of service in terms of delay and bandwidth to sessions with the same weight depending on their traffic pattern. To minimize such long-term unfairness, we propose a delay-bandwidth normalization model that defines the concept of value of service (VoS) from the aspect of both delay and bandwidth. A model and a packet-by-packet scheduling algorithm are proposed to realize the VoS concept. Performance comparisons between the proposed algorithm and algorithms based on fair queueing and service curve show that the proposed algorithm provides better long-term fairness among sessions and that is more adaptive to dynamic traffic characteristics without compromising its service rate and delay guarantees.

• The KIPS Transactions:PartC
• /
• v.9C no.1
• /
• pp.123-134
• /
• 2002

The requirement for QoS (Quality of Service) has become an important Issue as real-time or high bandwidth services are increasing, such as Internet Telephony, Internet broadcasting, and multimedia service etc. In order to guarantee the QoS of Internet application services, several approaches are being sought including IntServ (Integrated Service) DiffServ(Differentiated Srvices), and MPLS(Multi-Protocol Label Switching). In this paper, we describe the performance analysis of QoS guarantee mechanism using the DiffServ. To analyze how the DiffServ performance was affected by diverse input traffic models and the weight value in WFQ(Weighted Fair Queueing), we simulated and performed performance evaluation under a random, bursty, and self-similar input traffic models and for diverse input parameters. leased on the results of performance analysis, it was confirmed that significant difference exist in packet delay and loss depending on the input traffic models used. However, it was revealed that QoS guarantee is possible to the EF (expedited Forwarding) class and the service separation between RF and BE (Best Effort) classes may also be achieved. Next, we discussed the performance synthesis problem. (i. e. derived the conservation laws for a DiffServ networks, and analysed the performance variation and dynamic behavior based on the resource allocation (i.e., weight value) in WFQ.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.4B
• /
• pp.692-699
• /
• 2000

We build and analyze some types of queueing model to discuss capacity dimensioning of access links of a LAN interworking unit connected to the Internet backbone network. We assume that the IWU has a FIFO buffer to transmit IP packets to the Internet through the backbone. In order to analyze the system, we use a Poisson process and an MMPP process as input traffic models of IP packets and we use a general service time distribution as a service time model. But we use both an exponential service time and a deterministic service time in numerical examples for simple and efficient performance comparisons. As performance measures, we obtain the packet loss probability and the mean packet delay. We present some numerical results to show the effect of arrival rate, buffer size and link capacity on packet loss and mean delay.

• ETRI Journal
• /
• v.36 no.3
• /
• pp.396-406
• /
• 2014

We have designed and investigated a new congestion control scheme, called optimal and fully explicit (OFEX) controller. Different from existing relatively explicit controllers, this new scheme can provide not only optimal bandwidth allocation but also a fully explicit congestion signal to sources. It uses the congestion signal from the most congested link instead of the cumulative signal from the flow path. In this way, it overcomes the drawback of relatively explicit controllers exhibiting bias toward multi-bottlenecked users and significantly improves their convergence speed and source throughput performance. Furthermore, our OFEX-controller design considers a dynamic model by proposing a remedial measure against the unpredictable bandwidth changes in contention-based multi-access networks. Compared with former works/controllers, this remedy also effectively reduces the instantaneous queue size in a router and thus significantly improves queuing delay and packet loss performance. We have evaluated the effectiveness of the OFEX controller in OPNET. The experimental comparison with the existing relatively explicit controllers verifies the superiority of our new scheme.

• Journal of The Institute of Information and Telecommunication Facilities Engineering
• /
• v.1 no.1
• /
• pp.51-64
• /
• 2002

This paper analyzes performance measures of a Bluetooth_based mobile home network system. The home network system consists of terminals with Bluetooth interfaces, access points (AP), a home PC, and a gateway A mobile host in wireless terminals uses Mobile IP for supporting the mobility This paper considers four types of data traffic, which are new connection traffic, handoff traffic, Internet data traffic, and control data traffic and suggests a queueing system model of the home network system, where the AP and the home PC are modeled as M/G/1 with four priority queues and the gateway is modeled as M/G/1 with a single queue The generation rate and service time of individual traffic influence their performance measures. Based ell the suggested model, we propose the elapsed time of data traffic in terms of the number of cells, the number of Home PCs, arrival rates of four types of traffic and the service rates of AP/Home PCs/Gateway To analyze influences on the elapsed time with respect to arrival rate of four types of traffic, some examples are given.

• Annual Conference of KIPS
• /
• 2001.10b
• /
• pp.1533-1536
• /
• 2001

This paper considers the TMN system for management of public ATM switching network which has the four-level hierarchical structure consisting of one network management system, a few element management system and several agent-ATM switch pairs, respectively. The effects of one TMN command on the local call processing performance of the component ATM switch an analyzed using Jackson's queueing model. The TMN command considered is the permanent virtual call connection, and the performance measures of ATM switch are the utilization, mean queue length and mean waiting time for the processor interfacing the subscriber lines and trunks directly, and the call setup delay of the ATM switch.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.24 no.3
• /
• pp.83-91
• /
• 1999

In order to build and manage an ATM network effectively under several types of control methods, it is necessary to estimate the performance of the equipments in various viewpoints, especially of ATM multiplexer. As for the method to model the input stream into the ATM multiplexer, many researches have been done to characterize it by, such as, fluid flow, MMPP(Markov Modulated Poisson Process), or MMDP (Markov Modulated Deterministic Process). We introduce an MRP(Markov Renewal Process) to model the input stream which has proper structure to represent the burst traffic with high correlation. In this paper, we build a model for aggregated heterogeneous ON-OFF sources of ATM traffic by MRP. We make discrete time MR/D/1/B queueing system, whose input process is the superposed MRP and present a performance analysis by finding CLP(Cell Loss Probability). A simulation is done to validate our algorithm.

• IE interfaces
• /
• v.25 no.4
• /
• pp.431-440
• /
• 2012

Since pooling is a popular scheme in many areas to attain operational excellence, many researchers investigated the performance of pooling systems. However, rare research could be found on pooling with game situation which has much applicability to real world phenomenon. We analyze the performance of noncooperative pooling system with two servers having different sharing capacity. We investigate the sensitivity of the advantage of capacity pooling on the variation of system parameters, including sharing capacity numbers, pooling probability, pooling strategy and traffic intensity. As a result, we suggest an efficient control policy which facilitate the performance of pooling in a game situation.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.22 no.2
• /
• pp.99-123
• /
• 1997

Performance analysis of NTT/BT protocol is investigated, which is a GFC (Generic Flow Control) ptotocol in ATM (Asynchronous Transfer Mode ) network and is based on cyclic reset mechanism. THe mean cell delay time is proposed as a performance measure of NTT/BT protocol. The mean cell delay time is defined as the duration from the instant the cell arrives at the transmission buffer until the cell is fully transmitted. The process of cell transmission can be described as a single server queueing modle with two dependent services. By utilizing this model, mean cell delay time is obtained and sensitivity of the factors such as window size and reset period is also analysed.

• Journal of the Korean Statistical Society
• /
• v.35 no.3
• /
• pp.281-303
• /
• 2006

In an earlier work, we investigated the problem of using linear programming to bound performance measures in a discrete time Jackson network. There it was assumed that the system evolution is controlled by the early arrival scheme. This assumption implies that the system can't be modelled by a Markov chain. This problem was resolved and performance bounds were calculated. In the present work, we use a modification of the early arrival scheme (without corrupting it) in order to make the system evolves as a Markov chain. This modification enables us to obtain explicit expressions for certain moments that could not be calculated explicitly in the pure early arrival scheme setting. Moreover, this feature implies a reduction in the linear program size as well as the computation time. In addition, we obtained tighter bounds than those appeared before due to the new setting.