• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.7
• /
• pp.813-826
• /
• 1999

실시간 멀티미디어 응용의 등장으로 멀티캐스트와 QoS(Quality of Service) 지원이 필수적인 망 서비스로 부각되고 있다. 이에, ATM 기반의 인터넷에서 IP 멀티캐스트의 효율적인 처리를 위하여 MARS(Multicast Address Resolution Server)가 제안되었고, 기존의 최선 서비스 기반의 인터넷에서 QoS(Quality of Service)를 지원하기 위하여 RSVP(Resource Reservation Protocol)가 제안되었다. 본 논문에서는 ATM 망에서 QoS가 지원되는 IP 멀티캐스트 서비스를 제공하기 위하여 MARS 구조에서 RSVP를 지원하는 두 가지 방안을 제안하고, 시뮬레이션을 통하여 그 성능을 분석하였다. 제안하는 두 가지 방법은 각각 'RSVP 전 홉 노드 방식'과 'MARS 서버 방식'이라 명명하였다. RSVP 전 홉 노드 방식은 송신원으로부터 ATM 망으로 진입하는 노드와 수신원을 향하여 ATM 망을 진출하는 노드 간에 각각 일대일 양방향 VC를 설정하여 멀티캐스트 그룹에 속하는 수신원들이 보내는 자원 예약 메시지를 ATM 망에서 전송하는 방안이다. MARS 서버 방식은 ATM 망을 진출하는 노드와 MARS 서버간에 MARS 제어 메시지 교환을 위해 존재하는 ATM VC를 사용하여 RSVP의 자원 예약 메시지를 전송하고, MARS 서버가 RSVP 자원 예약 메시지를 처리하도록 그 기능을 확장함으로써 ATM 망에서 필요로 하는 제어 VC 수를 절약할 수 있는 방안이다. 시뮬레이션을 통하여, MARS 서버 방식은 ATM 제어 VC의 수를 절약할 수 있을 뿐 아니라 경우에 따라 RSVP 자원 예약 메시지 전달 지연을 줄일 수도 있음을 볼 수 있었다. 그러나, MARS 클러스터 내에 동시에 존재하는 RSVP 흐름이 많을 때에는 MARS 서버 방식의 경우 MARS 서버에서의 병목 현상으로 인해 성능이 저하될 수 있다.Abstract Emerging real time multimedia applications require multicast service with a QoS(Quality of Service) support. An overlay service architecture MARS(Multicast Address Resolution Server) is proposed to support IP multicast over an ATM network, and a resource reservation protocol RSVP is proposed to provide QoS support in the Internet which is originally based upon best effort service only. In this paper, we propose two schemes to support IP multicast service with QoS support over ATM networks: 'RSVP Previous Hop Node(RPHN) scheme' and 'MARS server based scheme'. In RPHN scheme, the RSVP reservation messages are transported via one-to-one ATM control VC from the egress nodes to the ingress nodes of the the multicast flow set up between each pair of nodes. The RSVP message processing occurs at the ingress nodes of the multicast flow. Whereas, in the MARS server based scheme, the RSVP reservation messages are transported via the MARS control VCs between the egress nodes and the MARS server. The RSVP message processing burden is imposed at MARS server in this scheme. For MARS server based scheme, no additional ATM VC is required for RSVP reservation message transmission, while the processing burden at the MARS server is high. Simulation results show that the MARS server based scheme, may accomplish RSVP reservation message delivery with smaller delay as well as saving of the number of ATM VCs. When the number of simultaneous RSVP flows in the MARS cluster is large, however, MARS based scheme may suffer performance degradation since MARS server becomes a performance bottleneck.

• The KIPS Transactions:PartA
• /
• v.14A no.1 s.105
• /
• pp.63-72
• /
• 2007

This paper introduces a load-balancing algorithm focusing on distributing web requests evenly into the web cluster servers. The load-balancing algorithms based on conventional periodic load-information update mechanism are not suitable for dynamic page applications, which are common in Cyber Foraging services, due to the problems caused by periodic synchronized load-information updating and the difficulties of work load estimation caused by embedded executing scripts of dynamic pages. Update-on-Finish algorithm solves this problem by using non-periodic load-update mechanism, and the web switch knows the servers' real load information only after their reporting and then distributes new loads according to the new load-information table, however it results in much communication overhead. Our proposed mechanism improve update-on-finish algorithm by using K-Percents-Finish mechanism and thus largely reduce the communication overhead. Furthermore, we consider the different capabilities of servers with a threshold Ti value and propose a load-balancing algorithm for servers with various capabilities. Simulation results show that the proposed K-Percents-Finish Reporting mechanism can at least reduce 50% communication overhead than update-on-finish approach while sustaining better load balancing performance than periodic mechanisms in related work.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.6B
• /
• pp.1113-1122
• /
• 1999

In this paper, we propose MSS(Multicast Synchronization Server) and QCS(QoS Conversion Server) models that can support IP multicast and QoS(Quality of Service) over the ATM network more efficiently. In the MSS model, it is possible to establish shortcut VCs(Virtual Circuits) among all hosts in the ATM network and to transfer multicast data at high speed. Also the MSS model is more scalable, because the number of inter-cluster VCs needed in the MSS model is less than that of EARTH. In the QCS model, ATM switch is modified to support one QoS service and best effort service through the one point-to-multipoint VC at the same time so required network resources are reduced, and dedicated server is used for QoS conversion to accept heterogeneous receivers more efficiently. In this Paper, the proposed MSS model and QCS model have solved both efficiency problem and scalability problem. It is proved through the comparison of the VCs required in each model.

• Journal of Korea Multimedia Society
• /
• v.14 no.6
• /
• pp.835-848
• /
• 2011

In this paper, we propose a client server based distributed/parallel system to effectively visualize huge geographic data. The system consists of a web-based client GUI program and a distributed/parallel server program which runs on multiple PC clusters. To make the client program run on mobile devices as well as PCs, the graphical user interface has been designed by using JOGL, the java-based OpenGL graphics library, and sending the information about current available memory space and maximum display resolution the server can minimize the amount of tasks. PC clusters used to play the role of the server access requested geographic data from distributed disks, and properly re-sample them, then send the results back to the client. To minimize the latency happened in repeatedly access the distributed stored geography data, cache data structures have been maintained in both every nodes of the server and the client.

• The KIPS Transactions:PartA
• /
• v.8A no.3
• /
• pp.217-226
• /
• 2001

In this paper, for continuous media access operations performed by Clustered Continuous Media Storage Server (CCMSS) system, we present the analytical model based on the open queueing network, which considers simultaneously two critical delay factors, the disk I/O and the internal network, in the CCMSS system. And we derive by using the analytical model the stochastic model for the total service delay time in the system. Next, we propose the integrated stochastic admission control model for the CCMSS system, which estimate the maximum number of admittable service requests at the allowable service failure rate by using the derived stochastic model and apply the derived number of requests in the admission control operation. For the performance evaluation of the proposed model, we evaluated the deadline miss rates by means of the previous stochastic model considering only the disk I/O and the propose stochastic model considering the disk I/O and the internal network, and compared the values with the results obtained from the simulation under the real cluster-based distributed media server environment. The evaluation showed that the proposed admission control policy reflects more precisely the delay factors in the CCMSS system.

• Journal of Digital Convergence
• /
• v.6 no.2
• /
• pp.117-125
• /
• 2008

High-volume web sites often use clusters of servers with load balancing as a way to increase the performance, scalability, and availability of the sites. Load balancing, usually performed by load balancer in front of such clusters, is a technique to spread workload between several computers or resources, in order to get optimal resource utilization or response time. In this paper we examine the performance for several configurations of cluster-based web servers using a simulation approach. We investigate two types of buffering scheme (common and local) for web clusters and three load balancing policies (uniformly random, round robin, and least queue first), using response time as a performance measure. We also examine two basic approaches of scaling web clusters: adding more servers of same type or upgrading the capacity of the servers in the clusters.

• The KIPS Transactions:PartA
• /
• v.12A no.5 s.95
• /
• pp.427-432
• /
• 2005

I studied about existent load distribution algorithms and the WLC(Weighted Least Connection) algerian that is using much at present to distribute the connection request of users to real servers efficiently in web cluster system. The efficiency of web cluster system is fallen by load imbalance between servers, because there is problem In inaccurate load status measuring of servers and measuring timing at these load distribution algorithms. In this paper, I suggest an algorithm that distributes load base on various load state of servers by real time using broadcasting RPC(Remote Procedure Call) when a user requests connection, and implement a prototype and experiment its performance. The experiment result shows that load imbalance phenomenon between reai sowers was improved greatly than existing method, and performance of web cluster system was improved by efficiency that response time is shortened.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.11
• /
• pp.1041-1045
• /
• 2010

Clusters can provide scalable and flexible architectures for parallel computing servers and data centers. Their performance prediction has been a very challenging issue. Existing performance measurement methodologies are able to measure the performance of servers already constructed. Thus they cannot provide a way to predict the overall system performance in advance when designing the system at the initial phase or adding more nodes for more capacity. Therefore, the performance modeling and prediction methodology for large-scale clusters is highly required. In this paper, we suggest a methodology to predict the performance of large-scale clusters, which consists of measurement, modeling and prediction steps. We apply the methodology to a real cluster server and show its usefulness.

• International journal of advanced smart convergence
• /
• v.10 no.4
• /
• pp.278-288
• /
• 2021

Recently, IoT systems are cloud-based, so that continuous and large amounts of data collected from sensor nodes are processed in the data server through the cloud. However, in the centralized configuration of large-scale cloud computing, computational processing must be performed at a physical location where data collection and processing take place, and the need for edge computers to reduce the network load of the cloud system is gradually expanding. In this paper, a cluster system consisting of 6 inexpensive Raspberry Pi boards was constructed to perform fast data processing. And we propose "Kubernetes cluster system(KCS)" for processing large data collection and analysis by model distribution and data pipeline method. To compare the performance of this study, an ensemble model of deep learning was built, and the accuracy, processing performance, and processing time through the proposed KCS system and model distribution were compared and analyzed. As a result, the ensemble model was excellent in accuracy, but the KCS implemented as a data pipeline proved to be superior in processing speed..

• The Plant Pathology Journal
• /
• v.37 no.4
• /
• pp.389-395
• /
• 2021

Soil is the major source of plant-associated microbes. Several fungal and bacterial species live within plant tissues. Actinomycetes are well known for producing a variety of antibiotics, and they contribute to improving plant health. In our previous report, Streptomyces globisporus SP6C4 colonized plant tissues and was able to move to other tissues from the initially colonized ones. This strain has excellent antifungal and antibacterial activities and provides a suppressive effect upon various plant diseases. Here, we report the genome-wide analysis of antibiotic producing genes in S. globisporus SP6C4. A total of 15 secondary metabolite biosynthetic gene clusters were predicted using antiSMASH. We used the CRISPR/Cas9 mutagenesis system, and each biosynthetic gene was predicted via protein basic local alignment search tool (BLAST) and rapid annotation using subsystems technology (RAST) server. Three gene clusters were shown to exhibit antifungal or antibacterial activity, viz. cluster 16 (lasso peptide), cluster 17 (thiopeptide-lantipeptide), and cluster 20 (lantipeptide). The results of the current study showed that SP6C4 has a variety of antimicrobial activities, and this strain is beneficial in agriculture.