• Journal of Web Engineering
• /
• v.21 no.1
• /
• pp.27-52
• /
• 2021

With recent increases in the number of network-connected devices, the number of edge computing services that provide similar functions has increased. Therefore, it is important to recommend an optimal edge computing service, based on quality-of-service (QoS). However, in the real world, there is a cold-start problem in QoS data: highly sparse invocation. Therefore, it is difficult to recommend a suitable service to the user. Deep learning techniques were applied to address this problem, or context information was used to extract deep features between users and services. However, edge computing environment has not been considered in previous studies. Our goal is to predict the QoS values in real edge computing environments with improved accuracy. To this end, we propose a GAIN-QoS technique. It clusters services based on their location information, calculates the distance between services and users in each cluster, and brings the QoS values of users within a certain distance. We apply a Generative Adversarial Imputation Nets (GAIN) model and perform QoS prediction based on this reconstructed user service invocation matrix. When the density is low, GAIN-QoS shows superior performance to other techniques. In addition, the distance between the service and user slightly affects performance. Thus, compared to other methods, the proposed method can significantly improve the accuracy of QoS prediction for edge computing, which suffers from cold-start problem.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.3 no.4
• /
• pp.198-206
• /
• 2008

Internet protocol television (IPTV), one of the most emerging services, offers multimedia streaming services with security, reliability, and relevant quality of service (QoS)/quality of experience (QoE). It provides added values to all the involving players and also brings technical and business challenges to them. For IPTV services, we expect to adopt the next generation network (NGN) environment for high quality and the Web technologies for personalization to meet the customer's necessity. Web can provide an open, flexible, and agile platform. Therefore, in this paper, we propose personalized IPTV services based on Web-based open platform in NGN environment and present functional architecture for Web-based personalized IPTV services. Technical issues for deploying the proposed IPTV services using Web are also provided. The objective of this paper is to analyze the critical architectural issues for developing viable and feasible networking platform models for personalized IPTV services.

• Journal of KIISE:Software and Applications
• /
• v.36 no.1
• /
• pp.1-9
• /
• 2009

In heterogeneous and distributed computing environments, with an increasing number of Web services providing similar functionalities, the reliability of Web services is a critical decision factor. To fulfill the open business model such as cooperation among enterprises, several Web services can be composed into the upper level business transaction. In Web services composition, the reliability of services is more and more critical. Though each unit Web service can be reliable, the reliability of the composed service is not guaranteed. Thus a way to efficiently assess and select composed Web services is needed. In this paper, we define new metrics for measuring the trust value of Web services, and propose an evaluation method to predict the trustworthy degree of the composed services based on the metrics. We also define a conceptual framework to support optimal Web services selection based on the proposed trust evaluation method. By selecting using the quantitative measurement rather than intuitive selection of the service user, it allows the service users to select the high reliable service meeting their quality requirements well.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1998.11a
• /
• pp.370-373
• /
• 1998

현재 인터넷상의 QoS(Quality of Service) 향상을 위해서 여러 가지 방법과 기술들이 소개되고 있다. 이러한 QoS 향상을 위해 트래픽 제어기법과 IS(Integrated Services), DS(Differentiated Services), Tag Switching, MPLS(Multi protocol Label Switching) 등과 같은 방법으로 인터넷에서 보다 나은 서비스를 지원하려는 노력이 시도되고 있다. 본 논문에서는 특히 TCP connection에서의 initial window size의 변경이 네트워크의 QoS, 즉 link의 utilization과 packet의 drop, performance에 어떠한 영향을 주는지를 살펴본다. 그리고, 이에 대한 특성을 분석하고 개선점을 현재의 인터넷에서 가장 많이 사용되어지고 있는 FTP와 web에 대한 모델을 사용하여 가상적으로 시뮬레이션을 해본다.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2980-2982
• /
• 2000

본 연구에서는 리눅스 기반의 클러스터링 웹서버를 구성하고, 이 클러스터링 웹서버를 하나의 서버인 것처럼 관리하는 개발툴을 만들었다. 그리고 커널 패치를 통하여 로드밸런서가 다양한 시스템 정보를 밸런싱에 이용하도록 하였다. 각 리얼서버에는 응답 데이터의 양 혹은 각 패킷의 길이에 따라 전송의 순서를 결정하는 QoS를 구현하였다.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 2004.07a
• /
• pp.285-285
• /
• 2004

• The Journal of the Korea Contents Association
• /
• v.12 no.4
• /
• pp.69-75
• /
• 2012

In this paper, to provide various services of QoS, and moreover applying traffic ratio to CAC(Call Admission Control) algorithm tested how long average data rate and the average packet delay time. When CAC algorithm is not applied, traffic mixture ratio is 1:1:4:4, the FTP Service 0.4, web services 0.6, streaming service 0.7, the packet delay requirements are not satisfied. On the other hand CAC Algorithm is applied, all the service of packet delay are satisfied with arrival rate. Therefore, we can make sure that applying of CAC of traffic control WWW, FTP, Video, VoIP can guarantee the various services of QoS.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06c
• /
• pp.7-9
• /
• 2006

본 논문은 워크플로우 분할에 기반하여 복합 웹 서비스를 빠르게 선택하는 방법을 제안한다. 제안된 방법은 QoS 요구사항의 제약정도 계산, 워크플로우 분할, QoS 요구사항 분해, 그리고 서비스 선택의 네 단계로 구성된다. 특히 QoS 요구사항의 분해과정이 휴리스틱에 기반하기 때문에 워크플로우 분할 시 QoS 요구사항을 만족하는 서비스를 찾지 못할 가능성이 있다. 제안된 방법은 분할에 따른 서비스 선택 실패를 줄이기 위해 주어진 QoS 요구사항의 제약정도를 계산하여 워크플로우의 분할 여부를 판단한다. 실험 결과 제안된 방법은 분할 속도, 성공률, 품질 면에서 우수한 성능을 보였다.

• Journal of the Korea Society for Simulation
• /
• v.14 no.3
• /
• pp.119-127
• /
• 2005

Web services and their composition (web processes) are promising technologies to efficiently integrate disparate software components over various types of systems. As many web services are nowadays available on Internet, quality of services (QoS) and performance/cost become increasingly important to differentiating between similar service providers. In this work, we introduce sPAC (Web Services Performance Analysis Centre) and show how customers can benefit from sPAC to consider performance in composing and commercializing web services. sPAC 1) helps users to graphically describe the workflow of web services, 2) invokes web services to test out performance for light load conditions, 3) automatically converts the web services and the flow between them into a simulation model, 4) conducts extensive simulations for heavy load conditions and various usage patterns, and 5) reports analysis results and estimation data for the web services.

• Journal of Korea Multimedia Society
• /
• v.11 no.3
• /
• pp.341-352
• /
• 2008

The dedicated line linked up with the internet maintains generally the efficiency of network by using Best Effort Model. But it is the problem that we allocate the higher bandwidth to services which request high bandwidth regardless of service types or purposes. To overcome this weak point, we use the Session QoS System which distinguish between business services such as E-mail, Web, etc. and non-business services. But, a recent session QoS system is not adequate to the environment similar to university, because that has many users who have various requests. In this paper, to satisfy various requirements of many user's in session QoS system which support services of research, education, business, etc., we analyze types of service in the environment of education and research network and propose a necessity of management strategies of session Qos system.