• The KIPS Transactions:PartC
• /
• v.16C no.4
• /
• pp.417-422
• /
• 2009

Implementation of DPI(Deep Packet Inspection) system on a general purpose multiprocessor platform is an attractive option from the implementation cost point of view, since it does not require high-cost customized hardware. Load balancing has been considered as a primary means to achieve high performance in multi processor systems. We claim, however, that in case of DPI system design simply balancing the load of each processor does not necessarily yield the highest system performance. Instead, we propose a method in which tasks are allocated to processors based on their functions. We implemented the proposed method in dual processor Linux system and compare its performance with the existing load balancing methods. Under the proposed method, one processor is dedicated to deal with interrupt handling and generic packet processing, while another processor is dedicated to DPI processing. According to experimental results, the proposed scheme outperforms the existing schemes by 60%, mainly because of the reduction of cache miss and spin lock occurrences.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.12
• /
• pp.437-444
• /
• 2022

Recently, research on mobile edge services has been conducted to handle computationally intensive and latency-sensitive tasks occurring in wireless networks. However, MEC, which is fixed on the ground, cannot flexibly cope with situations where task processing requests increase sharply, such as commuting time. To solve this problem, a technology that provides edge services using UAVs (Unmanned Aerial Vehicles) has emerged. Unlike ground MEC servers, UAVs have limited battery capacity, so it is necessary to optimize energy efficiency through load balancing between UAV MEC servers. Therefore, in this paper, we propose a load balancing technique with consideration of the energy state of UAVs and the mobility of vehicles. The proposed technique is composed of task offloading scheme using genetic algorithm and task migration scheme using Q-learning. To evaluate the performance of the proposed technique, experiments were conducted with varying mobility speed and number of vehicles, and performance was analyzed in terms of load variance, energy consumption, communication overhead, and delay constraint satisfaction rate.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06d
• /
• pp.190-192
• /
• 2006

모바일 기기 시장의 폭발적인 성장과 그에 따른 서비스의 수요증가로 인해 기존 유선 환경에서 제공되던 서비스를 모바일 환경에서도 그대로 적용하고자 하는 연구가 진행되고 있다. 그러나 멀티미디어와 같이 대량의 데이터를 필요로 하는 서비스 및 대용량 데이터베이스와 같이 블록 단위의 입출력을 필요로 하는 서비스를 한정된 저장 공간과 무선 환경이라는 제한된 특성을 가진 모바일 기기에서 제공하기에는 많은 제약이 따르게 된다. 본 논문에서는 Multi-Connection[1]을 지원하는 모바일 기기의 대용량 데이터 전송에 있어 연결된 Connection이 끊기는 경우, 복구를 수행하는 동안 다른 Connection들을 통해 전송하려는 데이터들을 효율적으로 분산하여 전송하기 위한 iSCSI 기반의 원격 스토리지 시스템[2]의 로드 밸런싱 기법에 대해 기술한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.1
• /
• pp.63-71
• /
• 2014

In this paper we researched on the Duplication Techniques for Active-Active Web Servers. Rsync and crontab utilities make copy periodically between web servers and maintain the same status. Load Balancing Server makes web servers load balanced and fast servicing by executing web servers alternatively. Even though one web server stops due to some critical errors, the remaining web server can take over and provide services continuously.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.214-216
• /
• 2001

웹 GIS에서의 인터넷 서비스 이용자의 집중 현상으로 발생하는 서버의 과부하 현상을 막고 안정적인 서비스 제공을 위해서는 웹 클러스터링 기술의 도입이 필요하다. 공간 질의는 웹 데이터와는 달리 인접 영역에 대한 질의가 매우 잣으며, 질의 처리 결과가 대용량이라는 특성을 가지고 있다. 이러한 공간 질의의 특성을 고려하지 않을 경우, 서버에서 처리되는 질의들의 지역적 인접성이 낮아지고 낮은 버퍼 재 사용율은 디스크로의 접근 빈도를 높여 질의 처리 비용을 증가시키는 원인이 된다. 본 논문에서는 웹 GIS 클러스터링 시스템을 위한 질의 영역의 인접성을 이용한 로드 밸런싱 기법을 제안한다. 제안하는 기법은 공간 데이터를 타일을 기반으로 인접한 타일의 그룹을 생성하여 각 서버에 할당하여, 질의 영역 및 공간 연산을 고려하여 서버에서 질의가 처리되는 동안 버퍼 재사용율이 최대가 되도록 클라이언트의 질의 요청을 적절한 서버에 전달한다. 제안하는 기법은 서버의 버퍼를 공간 인덱스 탐색에 최적화함으로써 서버의 버퍼 재상용율을 높이고, 클러스터링 시스템에서 디스크의 접근 횟수를 낮추어, 전체적인 서버 시스템의 처리 능력을 형상시킨다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.200-202
• /
• 2015

본 논문에서는 HEVC 타일 병렬 부호화시 타일간 로드 밸런싱을 통해 부호화 속도를 향상시키는 방법을 제안한다. HEVC 는 병렬 부호화를 통한 부호화기 속도 향상을 위해 타일 단위의 부호화 기법을 제공한다. 하지만, 타일로 나누어진 영역의 부호화 시간 차이는 병렬 부호화를 통한 속도 개선 효과를 감소시킨다. 제안된 방법은 타일간 부호화 시간차이를 고려하여 부호화 옵션을 다르게 적용함으로써 부호화 효율 감소대비 부호화 속도 성능을 개선하였다.

• Journal of the Korea Society for Simulation
• /
• v.19 no.2
• /
• pp.81-89
• /
• 2010

Biological data mining has been noticed as an issue as the volume of biological data is increasing extremely. Grid technology can share and utilize computing data and resources. In this paper, we propose a hybrid system that combines biological data mining with grid technology. Especially, we propose a decision range adjustment algorithm for processing efficiency of biological data mining. We obtain a reliable data mining recognition rate automatically and rapidly through this algorithm. And communication loads and resource allocation are key issues in grid environment because the resources are geographically distributed and interacted with themselves. Therefore, we propose a dynamic load balancing algorithm and apply it to the grid-based biological data mining method. For performance evaluation, we measure average processing time, average communication time, and average resource utilization. Experimental results show that this method provides many advantages in aspects of processing time and cost.

• Journal of Digital Contents Society
• /
• v.17 no.6
• /
• pp.581-592
• /
• 2016

WebRTC has quickly grown to be the world's advanced real-time communication in several platforms such as web and mobile. In spite of the advantage, the current technology in WebRTC does not handle a big-streaming efficiently between peers and a large amount request of users on the Signaling server. Therefore, in this paper, we put our work to handle the problem by delivering the flow of data with dynamical load balancing algorithms. We analyze the request source users and direct those streaming requests to a load balancing component. More specifically, the component determines an amount of the requested resource and available resource on the response server, then it delivers streaming data to the requesting user parallel or alternately. To show how the method works, we firstly demonstrate the load-balancing algorithm by using a network simulation tool OPNET, then, we seek to implement the method into an Ubuntu server. In addition, we compare the result of our work and the original implementation of WebRTC, it shows that the method performs efficiently and dynamically than the origin.

• KNOM Review
• /
• v.24 no.1
• /
• pp.1-12
• /
• 2021

VM (Virtual Machine) live migration is a server virtualization technique for deploying a running VM to another server node while minimizing downtime of a service the VM provides. Currently, in cloud data centers, VM live migration is widely used to apply load balancing on CPU workload and network traffic, to reduce electricity consumption by consolidating active VMs into specific location groups of servers, and to provide uninterrupted service during the maintenance of hardware and software update on servers. It is critical to use VMlive migration as a prevention or mitigation measure for possible failure when its indications are detected or predicted. In this paper, we propose two VNF live migration methods; one for predictive load balancing and the other for a proactive measure in failure. Both need machine learning models that learn periodic monitoring data of resource usage and logs from servers and VMs/VNFs. We apply the second method to a vEPC (Virtual Evolved Pakcet Core) failure scenario to provide a detailed case study.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.32 no.2
• /
• pp.38-46
• /
• 2004