• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권3호
• /
• pp.971-989
• /
• 2015

With the development of cloud computing, more and more Web services are deployed on geo-distributed datacenters and are offered to cloud users all over the world. Through service composition technologies, these independent fine-grain services can be integrated to value-added coarse-grain services. During the composition, a number of Web services may provide the same function but differ in performance. In addition, the distribution of cloud datacenters presents a geographically dispersive manner, which elevates the impact of the network on the QoS of composite services. So it is important to select an optimal composition path in terms of QoS when many functionally equivalent services are available. To achieve this objective, we first present a graph model that takes both QoS of Web services and QoS of network into consideration. Then, a novel approach aiming at selecting the optimal composition path that fulfills the user's end-to-end QoS requirements is provided. We evaluate our approach through simulation and compare our method with existing solutions. Results show that our approach significantly outperforms existing solutions in terms of optimality and scalability.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권7호
• /
• pp.2325-2349
• /
• 2014

Network virtualization technology plays a key role in cloud computing, which serves as an effective approach for provisioning a flexible and highly adaptable shared substrate network to satisfy the demands of various applications or services. Recently, the problem of mapping a virtual network (VN) onto a substrate network has been addressed by various algorithms. However, these algorithms are typically efficient for unicast service-oriented virtual networks, and generally not applicable to multicast service-oriented virtual networks (MVNs). Furthermore, the survivable MVN mapping (SMVNM) problem that considers the survivability of MVN has not been studied and is also the focus of this work. In this research, we discuss SMVNM problem under regional failures in the substrate network and propose an efficient algorithm for solving this problem. We first propose a framework and formulate the SMVNM problem with the objective of minimizing mapping cost by using mixed integer linear programming. Then we design an efficient heuristic to solve this problem and introduce several optimizations to achieve the better mapping solutions. We validate and evaluate our framework and algorithms by conducting extensive simulations on different realistic networks under various scenarios, and by comparing with existing approaches. Our simulation experiments and results show that our approach outperforms existing solutions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권7호
• /
• pp.3351-3369
• /
• 2017

In recent years, datacenters, network devices and computers have proliferated. The power consumed by information and communication technology (ICT) devices has inflated in an extraordinary manner. Green communication has emerged as a new approach to reduce and optimize power consumption in ICT sector. Many methods and protocols have been proposed and implemented to achieve green communication. Nevertheless, the increase of power consumption remains a problem. In this work, we attempt to reduce and optimize power consumption of network devices in datacenters environment utilizing software defined network (SDN) paradigm. To gain more insight of the power consumption requirements of network switches, a power measurement system is constructed to measure power consumption levels of network devices. Subsequently, we propose a duplication resolver algorithm (DRA) to power off/on switches reactively. DRA algorithm reduces the required time by switches to construct their flow tables after rebooting. To this end, DRA-based external circuit has been constructed utilizing Ethernet module and an Arduino kit to control power supplies of network devices. To facilitate our work, a testbed has been constructed utilizing Ryu SDN controller, HP2920-24G switches and Arduino kits. Our results show that DRA algorithm can reduce both the power usage and start-up time delay of network switches after failures.

• 정보과학회 논문지
• /
• 제44권7호
• /
• pp.719-725
• /
• 2017

약 네트워크의 관리 비용을 줄이고 성능을 향상시키기 위해 ETSI(European Telecommunication Standards Institute)는 클라우드 데이터 센터에서 네트워크 기능(Network Function)을 소프트웨어 형태로 구현할 수 있는 네트워크 기능 가상화(Network Function Virtualization) 개념을 도입했다. 네트워크 기능 가상화 구조 내에서 네트워크 기능을 물리적 노드(예: 범용 서버)에 네트워크 기능을 호스팅하여 실제 리소스를 공유할 수 있다. 네트워크 기능 가상화를 지원하는 네트워크 서비스 제공 업체의 경우, 효율적인 자원 할당 방법을 통해 운영비용(OPEX) 및 자본 비용(CAPEX)를 줄일 수 있다. 이에 본 논문에서는 최적화 방법을 통해 Network Service Chain Embedding 문제를 분석하고 Markov Approximation 프레임워크 기반 최적의 솔루션을 제안한다. 제안사항에 대한 시뮬레이션 결과는 평균 CPU 사용률이 73%, 링크 사용률이 최대 53% 증가함을 보여준다.

• 정보과학회 논문지
• /
• 제42권5호
• /
• pp.580-590
• /
• 2015

최근 그린 컴퓨팅의 일환으로 데이터센터의 효율적인 전력관리가 부각되고 있으며, 이러한 효율적인 데이터센터의 전력관리를 위해서는 서버 모니터링이 필수적이다. 하지만 기존 서버 모니터링 기법은 주로 데이터베이스만을 이용해 관리하였기 때문에, 관리자는 데이터센터 및 데이터베이스의 구조를 명확하게 알고 있어야 하고, 관측한 데이터 간 상관관계를 분석해야 한다. 또한 관리자는 데이터베이스에 명시되지 않은 부분에 대한 모니터링이 불가능하다. 이에 본 논문에서는 효율적인 데이터센터 내 전력관리를 위한 온톨로지 기반의 모니터링 기법을 제안한다. 제안된 기법은 서버관리를 위한 도메인 온톨로지를 구축하고, 구축한 온톨로지를 시맨틱 어노테이션 기술을 통해 데이터베이스와 연동하였으며, 서버의 상태 및 질의문 생성에 대한 규칙을 정의하였다. 그리고 제안하는 기법의 유효성 입증을 위해 온톨로지 기반 모니터링 시스템 아키텍처를 설계하고, 지식 체계를 구축하였으며, 파일럿 시스템을 구현하여 검증을 실시하였다.

• 전자통신동향분석
• /
• 제33권1호
• /
• pp.89-100
• /
• 2018

To accommodate the rapid increase in the amount of data traffic, the capacity of datacenters is expanding rapidly. Datacenter networks (DCNs) utilize electronic packet switches. However, the increases in the speed and capacity of electronic devices are slower than that of data expansion. Furthermore, electronics are too well developed to reach very near their physical limits. To achieve a breakthrough under this situation, optical switching schemes have been widely examined to replace or collaborate with incumbent electronic switches in a DCN. This report reviews the current status of such global researches on an optical DCN.

• ETRI Journal
• /
• 제38권3호
• /
• pp.529-539
• /
• 2016

Recently, a product-matrix (PM) framework was proposed to construct optimal regenerating codes for homogeneous distributed storage systems (DSSs). In this paper, we propose an extended PM (EPM) framework for coding of heterogeneous DSSs having different repair bandwidths but identical storage capacities. Based on the EPM framework, an explicit construction of minimum remote-repair bandwidth regenerating (MRBR) codes is presented for a specific heterogeneous DSS, where two geographically different datacenters with associated storage nodes are deployed. The data reconstruction and regeneration properties of the MRBR code are proved strictly. For the purpose of demonstration, an example implementation of MRBR code is provided. The presented MRBR code is the first optimal strict-regenerating code for heterogeneous DSSs. In addition, our proposed EPM framework can be applied to homogeneous systems also.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권12호
• /
• pp.4759-4775
• /
• 2015

The reduction of power consumption in large-scale datacenters is highly-dependent on the use of virtualization to consolidate multiple workloads. However, these consolidation strategies must also take into account additional important parameters such as performance, reliability, and profitability. Resolving these conflicting goals is often the major challenge encountered in the design of optimization strategies for cloud data centers. In this paper, we put forward a data center monitoring strategy which dynamically alters its approach depending on the cloud system's current state. Results show that our proposed scheme outperformed strategies which only focus on a single metric such as SLA-Awareness and Energy Efficiency.

• 설비공학논문집
• /
• 제22권8호
• /
• pp.530-537
• /
• 2010

With the advancement of technology, the density of IT equipment, heat load and power consumption continue to increase in high density internal-load dominated buildings as datacenters. To improve the HVAC system's energy performance and efficiency, there is a need to find methods of using outside air. Through cooling tower control that is based on outside wet-bulb temperature, the water-side economizer made it possible to achieve a maximum energy performance improvement of about 16.6% over the basic chilled water system, whereas the air-side economizer, through control based on outdoor air enthalpy, made it possible to achieve about 42.4% improvement.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2015년도 춘계학술발표대회
• /
• pp.6-9
• /
• 2015

유무선 네트워크 기술의 급격한 발전 및 개인용 휴대 장치의 보급 증가와 그에 맞물린 소셜네트워크 서비스 등의 활성화로 인해 데이터의 전송량이 폭발적으로 늘어나고 있으며, 이러한 추세는 향후 지속될 것으로 전망된다. 이에 데이터센터의 수요가 증가하고 있으며, 확장 및 유지보수가 쉬우면서도 높은 성능을 갖는 시스템에 대한 요구가 높아지고 있다. 본 논문에서는 데이터센터 네트워크의 구성요소를 분석하고 각각의 지연시간을 알아보았으며, 그 중 확장 및 유지보수가 쉬우나 상대적으로 지연시간이 높은 이더넷을 데이터센터에 적용하였을 때 지연시간 개선을 위한 방법으로 Intel DPDK를 적용할 경우, 미적용시와 비교하여 약 86%의 지연시간 감소를 확인하였고, 추가 향상 방안을 조망하였다.