• ETRI Journal
• /
• v.35 no.4
• /
• pp.687-696
• /
• 2013

The development of proxy caching is essential in the area of video-on-demand (VoD) to meet users' expectations. VoD requires high bandwidth and creates high traffic due to the nature of media. Many researchers have developed proxy caching models to reduce bandwidth consumption and traffic. Proxy caching keeps part of a media object to meet the viewing expectations of users without delay and provides interactive playback. If the caching is done continuously, the entire cache space will be exhausted at one stage. Hence, the proxy server must apply cache replacement policies to replace existing objects and allocate the cache space for the incoming objects. Researchers have developed many cache replacement policies by considering several parameters, such as recency, access frequency, cost of retrieval, and size of the object. In this paper, the Weighted-Rank Cache replacement Policy (WRCP) is proposed. This policy uses such parameters as access frequency, aging, and mean access gap ratio and such functions as size and cost of retrieval. The WRCP applies our previously developed proxy caching model, Hot-Point Proxy, at four levels of replacement, depending on the cache requirement. Simulation results show that the WRCP outperforms our earlier model, the Dual Cache Replacement Policy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.5
• /
• pp.1973-1991
• /
• 2018

Multimedia applications over wireless networks have been evolving to augmented reality or virtual reality services. However, a rich data size compared to conventional multimedia services causes bandwidth bottlenecks over wireless networks, which is one of the main reasons why those applications are not used widely. To overcome this limitation, bandwidth aggregation techniques, which exploit a multi-path transmission, have been considered to maximize link utilization. Currently, most of the conventional researches have been focusing on the user end problems to improve the quality of service (QoS) through optimal load distribution. In this paper, we address the joint pricing and load distribution problem for multi-homing in heterogeneous wireless access networks (ANs), considering the interests of both the users and the service providers. Specifically, we consider profit from resource allocation and cost of power consumption expenditure for operation as an utility of each service provider. Here, users decide how much to request the resource and how to split the resource over heterogeneous wireless ANs to minimize their cost while supporting the required QoS. Then, service providers compete with each other by setting the price to maximize their utilities over user reactions. We study the behaviors of users and service providers by analyzing their hierarchical decision-making process as a multileader-, multifollower Stackelberg game. We show that both the user and service provider strategies are closed form solutions. Finally, we discuss how the proposed scheme is well converged to equilibrium points.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.9
• /
• pp.4329-4348
• /
• 2019

Mobile cloud computing (MCC) can offload heavy computation from mobile devices onto nearby cloudlets or remote cloud to improve the performance as well as to save energy for these devices. Therefore, it is essential to consider how to achieve efficient computation offloading with constraints for multiple users. However, there are few works that aim at multi-objective problem for multiple users. Most existing works concentrate on only single objective optimization or aim to obtain a tradeoff solution for multiple objectives by simply setting weight values. In this paper, a multi-objective optimization model is built to minimize the average energy consumption, time and cost while satisfying the constraint of bandwidth. Furthermore, an improved multi-objective optimization algorithm called D-NSGA-II-ELS is presented to get Pareto solutions with better convergence and diversity. Compared to other existing works, the simulation results show that the proposed algorithm can achieve better performance in terms of energy consumption, time and cost while satisfying the constraint of the bandwidth.

• The KIPS Transactions:PartC
• /
• v.12C no.5 s.101
• /
• pp.695-700
• /
• 2005

To provide the fair rate and achieve the fair bandwidth allocation, many per-flow scheduling algorithms have been proposed such as fair queueing algorithm for congestion control. But these algorithms need to maintain the state, manage buffer and schedule packets on a per-flow basis; the complexity of these functions may prevent them from being cost-effectively implemented. In this paper, therefore, to acquire cost-effectively for implementation, we propose a CS-FNE(Core Stateless FNE) algorithm that is based on FM(Flow Number Estimation), and evaluated CS-FNE scheme together with CSFQ(Core Stateless Fair Queueing), FRED(Fair Random Early Detection), RED(Random Early Detection), and DRR(Dynamic Round Robin) in several different configurations and traffic sources. Through the simulation results, we showed that CS-FNE algorithm can allocate fair bandwidth approximately than other algorithms, and CS-FNE is simpler than many per-flow basis queueing mechanisms and it can be easily implemented.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.2
• /
• pp.177-187
• /
• 2000

In the past several years, many systems which adopted ring topology with high-speed unidirectional point-to-point links have emerged to overcome the limit of bus for interconnection network of clustered multiprocessor system. However, rapid increase of processor speed and performance improvement of local bus and memory system limit scalability of system with point-to-point link of standard bandwidth. Therefore, necessity to extend bandwidth is emphasized. In this paper, we adopt PANDA system as base model, which is clustering-based multiprocessor system. By simulating a model adopting commercial processor and local bus specification, we show that point-to-point link is bottleneck of system performance, and bandwidth expansion by more than 200% is needed. To expand bandwidth of interconnection network, it needs excessive design cost and time to develop new point-to-point link with doubled bandwidth. As an alternative to double bandwidth, we propose several ways to implement dual ring -simple dual ring, transaction-separated dual ring, direction-separated dual ring- by using off-the-shelf point-to-point links with IEEE standard bandwidth. We analyze pros. and cons. of each model compared with doubled-bandwidth single ring by simulation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.5B
• /
• pp.449-455
• /
• 2006

Due to advance in multimedia applications and integrated Internet services, the optical access networks have been actively studied. In particular, Ethernet passive optical network (EPON) has received much attention due to high bandwidth provision with low cost. In EPON system, the data transmission is carried out in two directions: downstream (from OLT to ONU) and upstream (from ONU to OLT). The downstream data is broadcasted to every ONUs, while the upstream data is point-to-point transmitted between each ONU and OLT, where the uplink is shared by all ONUs in the form of TDMA. The bandwidth allocation algorithm is required to efficiently manage the bandwidth on the uplink. The limited algorithm was proposed to enhance the capability of dynamic bandwidth allocation. In this paper, we propose the adaptive limited algorithm to enhance the shortcomings of limited algorithm. The adaptive limited algorithm enhances the dynamics on bandwidth allocation, and at the same time controls the fairness on packet delay. Through the computer simulations, it is shown that the adaptive limited algorithm achieves high dynamic on bandwidth allocation, maintains a good fairness on packet delay between ONUs, and keeps the fairness on the bandwidth on the demand basis.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.3
• /
• pp.77-85
• /
• 2009

Execution times of tasks can be variable depend on input data. If we choose a high performance processor to satisfy the worst case execution times, the hard cost becomes high and the energy consumption also becomes large. To apply a lower performance processor, we have to utilize processor capacity maximally while overrunning tasks can not affect deadlines of other tasks. To be used for such systems, this paper presents SBP (Shared Bandwidth Partitioning) that a processor bandwidth is reserved and shared among all tasks. If a task needs more processor capacity, it can use a portion of the shared bandwidth. A simulation result shows that SBP provides better performance than previous algorithms. SBP reduces deadline miss ratio which is related to scheduling quality. And the number of context switches, which is related to system overhead, is also reduced.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.11B
• /
• pp.891-899
• /
• 2004

Ethernet passive optical network (EPON) has drawn many attention as a promising access network technology because it can provide a high bandwidth with a low cost. Since the uplink in the EPON system is shared by many users, it is necessary for an EPON system to have an efficient bandwidth allocation mechanism. In addition, as the users demand more QoS-oriented applications, it is necessary for an EPON system to have an efficient mechanism supporting QoS. In this paper, we propose a variable window dynamic bandwidth allocation (DBA) algorithm for the EPON system. Unlike the previously proposed DBA algorithms, the variable window algorithm guarantees the QoS for the highest priority class, and at the same time provides more enhanced QoS for the lower priority classes by dynamically allocating bandwidth if necessary. It is verified through the simulations that the variable window algorithm can provide more enhanced QoS performance than other DBA algorithms.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.5
• /
• pp.143-150
• /
• 2019

Cloud computing has recently become an attractive solution for massively multiplayer online games(MMOGs), as it lifts operators from the burden of buying and maintaining hardware. Peer-to-peer(P2P) -based solutions present several advantages, including the inherent scalability, self-repairing, and natural load distribution capabilities. We propose a hybrid architecture for MMOGs that combines technological advantages of two different paradigms, P2P and cloud computing. An efficient and effective provisioning of resources and mapping of load are mandatory to realize an architecture that scales in economical cost and quality of service to large communities of users. As the number of simultaneous players keeps growing, the hybrid architecture relieves a lot of computational power and network traffic, the load on the servers in the cloud by exploiting the capacity of the peers. For MMOGs, besides server time, bandwidth costs represent a major expense when renting on-demand resources. Simulation results show that by controlling the amount of cloud and user-provided resource, the proposed hybrid architecture can reduce the bandwidth at the server while utilizing enough bandwidth of players.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.215-217
• /
• 2004

In wireless sensor networks, fair allocation of bandwidth among different nodes is one of the critical problems that effects the serviceability of the entire system. Fair bandwidth allocation mechanisms, like fair queuing, usually need to maintain state, manage buffers, and perform packet scheduling on a per flow basis, and this complexity may prevent them from being cost-effectively implemented and widely deployed. It is a very important and difficult technical issue to provide packet scheduling architecture for fairness in wireless sensor networks. In this paper, we propose an packet scheduling architecture for sensor node, called FISN (Fairness Improvement Sensor Network), that significantly reduces this implementation complexity yet still achieves approximately fair bandwidth allocations. Sensor node for sensing estimate the incoming rate of each sensor device and insert a label into each transmission packet header based on this estimate. Sensor node for forwarding maintain no per flow state; they use FIFO packet scheduling augmented by a probabilistic dropping algorithm that uses the packet labels and an estimate of the aggregate traffic at the gathering node. We present the detailed design, implementation, and evaluation of FISN using simulation. We discuss the fairness improvement and practical engineering challenges of implementing FISN in an experimental sensor network test bed based on ns-2.