• 정보과학회 논문지
• /
• 제43권1호
• /
• pp.27-33
• /
• 2016

오늘날 가상화 기술은 가장 유용하게 사용되는 컴퓨팅 기술 중 하나이며 모든 컴퓨팅 환경에서 널리 활용되고 있다. 그러나 가상화 환경의 I/O 계층들은 호스트 머신의 I/O 동작 방식을 인지하지 못하도록 설계되어 있기 때문에 I/O 확장성 문제는 여전히 해결해야 할 문제로 남아 있다. 특히, 다중 큐 SSD가 보조 기억 장치로 사용될 경우, 증가한 잠금 경쟁과 제한된 I/O 병렬화 문제로 가상 머신은 다중 큐 SSD의 공인된 성능을 활용하지 못하는 문제가 발생한다. 이러한 성능 문제를 해결하기 위해 본 논문에서는 가상 CPU마다 전용 큐와 I/O 스레드를 할당하는 새로운 기법을 제안한다. 제안 기법은 성능 저하의 주요한 원인 중 하나인 잠금 경쟁을 효율적으로 분산시키고 또 다른 원인인 Virtio-blk-data-plane의 병렬화 문제를 해소한다. 제안 기법을 평가한 결과 최신 QEMU 보다 IOPS가 최대 155% 향상되는 것을 확인하였다.

• 한국컴퓨터정보학회논문지
• /
• 제25권9호
• /
• pp.91-100
• /
• 2020

본 논문은 네트워크 품질에 대한 서비스의 품질과 성능을 평가할 수 있는 분산형/탈중앙 스토리지에 오디오와 비디오를 포함하는 멀티미디어 스트리밍 서비스의 프로토타입을 설계하고 구현한 통합된 평가 프레임워크를 제안한다. 본 논문의 평가 프레임워크는 기존 클라이언트·서버 구조의 확장성과 단일 장애 지점의 문제를 극복할 수 있도록 멀티미디어 스트리밍 서비스에 분산형 스토리지를 적용하였으며, 서비스 품질을 측정하고 향상 시키는 중요한 지표에 초점을 맞추었다. 특히 탈중앙 스토리지 기반의 멀티미디어 콘텐츠 스트리밍 서비스에서 스트리밍의 품질과 성능을 평가하기 위한 성능 지표를 측정할 수 있을 뿐만 아니라 네트워크 가상화 기술을 이용하여 네트워크 품질을 조정할 수 있다. 실험 결과에 따르면 제안된 프레임워크는 단일 호스트 컴퓨터에 탈중앙 스토리지 기반 스트리밍 서비스를 구축하고 운영하는 데 있어 오버헤드가 낮음을 보여주었으며, 평가 가능한 시스템 규모의 확장성을 검증하였다.

• 산업공학
• /
• 제24권2호
• /
• pp.139-150
• /
• 2011

The EPCglobal Network is a computer network used to share product data among trading partners. It provides the supply chain with improved visibility and traceability by using Electronic Product Code (EPC), which is stored on an RFID tag. Although this network model is widely accepted as a global standard and the growth of EPCglobal-subscriber base is considerable, the EPC technology adoption process is still in its infancy. This is because some of the critical issues on this model still remain to be verified such as scalability, data management, security, privacy and the economic value of data sharing. In this paper, we focus on scalability issue among the challenges to overcome and we regard performance of the EPCglobal Network only as a track and trace query-processing cost in the network. We developed performance models consisting of three elements of the EPCglobal Network : Discovery Services (DS), EPC Information Services (EPCIS), Object Naming Services (ONS). Then we abstracted out the track and trace query execution model to evaluate performance of the overall EPCglobal Network. Finally using the proposed models, we carried out simulation analysis based on an RFID-based inbound logistics process of automobile parts. This work is an important step towards the EPC technology diffusion and provides guidelines for businesses looking to buy or build the EPCglobal Network-based systems.

• 한국인터넷방송통신학회논문지
• /
• 제13권4호
• /
• pp.29-37
• /
• 2013

전형적으로 멀티플레이어 게임은 클라이언트-서버 구조 혹은 peer-to-peer 구조를 기반으로 구성된다. 클라이언트-서버 구조는 플레이어의 수가 증가할 때 서버에서 대역폭 요건이 커지기 때문에 규모조정이 어렵다. 반면 peer-to-peer 구조는 각 플레이어마다 자신의 상태와 다른 모든 플레이어들의 상태 간의 일관성을 점검해야 하기 때문에 상당한 오버헤드가 수반된다. 본 논문에서는 두 구조의 장점을 결합하는 기법을 제시한다. 이 기법에서는 플레이어들이 우선순위가 낮은 갱신 작업은 peer-to-peer 방식으로 상호 교환하며 우선순위가 높은 갱신 작업에 대해서는 중앙 서버와 직접 통신하게 된다. 결과적으로 제시된 기법은 클라이언트-서버 구조 보다 통신 부하 요건이 감소되며 서버의 대역폭 병목 현상도 제거된다. 멀티플레이어 게임의 다른 중요한 사항인 플레이어 간의 상태 일관성도 이 기법에서 항상 정확히 유지된다. 제시된 기법의 성능은 분석과 다양한 실험을 통하여 평가되었다.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2015년도 제49회 하계 정기학술대회 초록집
• /
• pp.232.2-232.2
• /
• 2015

Piezoelectric force microscopy (PFM) is a powerful method to characterize inversed piezoelectric effects directly using conductive atomic force microscopy (AFM) tips. Piezoelectric domains respond to an applied AC voltage with a characteristic strain via a contact between the tip and the surface of piezoelectric material. Electroactive piezoelectric polymers are widely investigated due to their advantages such as flexibility, light weight, and microactuation enabling various device features. Although piezoelectric polymers are promising materials for wide applications, they have the primary issue that the piezoelectric coefficient is much lower than that of piezoelectric ceramics. Researchers are studying widely to enhance the piezoelectric coefficient of the materials including nanoscale fabrication and copolymerization with some materials. In this report, nanoscale electroactive polymers are prepared by the electrospinning method that provides advantages of direct poling, scalability, and easy control. The main parameters of the electrospinning process such as distance, bias voltage, viscosity of the solution, and elasticity affects the piezoelectric coefficient and the nanoscale structures which are related to the phase of piezoelectric polymers. The characterization of such electroactive polymers are conducted using piezoelectric force microscopy (PFM). Their morphologies are characterized by field emission-scanning electron microscope (FE-SEM) and the crystallinity of the polymer is determined by X-ray diffractometer.

• Journal of Pharmaceutical Investigation
• /
• 제40권spc호
• /
• pp.9-17
• /
• 2010

Polymorphism has been recognized to be a critical issue throughout the drug product development process. Most of solid phase drugs have polymorphism, which has generated a great deal of interest and the field has been evolving rapidly. Preferably, thermodynamically most stable form of a drug substance is selected to obtain consistent bioavailability over its shelf life and various storage conditions. Moreover, it has the lowest potential for conversion from one polymorphic form to another. However, metastable or amorphous forms may be used intentionally to induce faster dissolution rate for rapid drug absorption and higher efficacy. For pharmaceutical industry, polymorphism is one of the key activities in form selection process together with salt selection. This article introduces the main features in the investigation of solid form selection especially polymorphic behavior with thermodynamic backgrounds, physicochemical properties with solubility, dissolution, and mechanical properties, and characterization techniques for proper analysis. The final form can be recommended based on the physicochemical and biopharmaceutical properties and by the processability, scalability and safety considerations. Pharmaceutical scientists especially in charge of formulation need to be well aware of the above issues to assure product quality.

• Journal of Computing Science and Engineering
• /
• 제9권4호
• /
• pp.163-176
• /
• 2015

Self-organization of distributed wireless sensor nodes is a critical issue in wireless sensor networks (WSNs), since each sensor node has limited energy, bandwidth, and scalability. These issues prevent sensor nodes from actively collaborating with the other types of sensor nodes deployed in a typical heterogeneous and somewhat hostile environment. The automated self-organization of a WSN becomes more challenging as the number of sensor nodes increases in the network. In this paper, we propose a dynamic self-organized architecture that combines tree topology with a drawn-grid algorithm to automate the self-organization process for WSNs. In order to make our proposed architecture scalable, we assume that all participating active sensor nodes are unaware of their primary locations. In particular, this paper presents two algorithms called active-tree and drawn-grid. The proposed active-tree algorithm uses a tree topology to assign node IDs and define different roles to each participating sensor node. On the other hand, the drawn-grid algorithm divides the sensor nodes into cells with respect to the radio coverage area and the specific roles assigned by the active-tree algorithm. Thus, both proposed algorithms collaborate with each other to automate the self-organizing process for WSNs. The numerical and simulation results demonstrate that the proposed dynamic architecture performs much better than a static architecture in terms of the self-organization of wireless sensor nodes and energy consumption.

• 정보통신설비학회논문지
• /
• 제3권1호
• /
• pp.15-24
• /
• 2004

'The delay and control overhead during route discovery for destinations outside ad hoc networks are major obstacle to achieving scalability in the Internet. To solve this issue, we propose a novel route discovery scheme equipped with two augmented functions. In this paper, the Internet gateway maintains an address cache of Internet nodes frequently accessed from the ad hoc network and replies with an extended Route Response (RREP) message to the Route Request (RREQ) message based on its routing table and the address cache called EXIT(EXternal node Information Table). These augmented functions make the source node determine the location of the destination as fast as possible. Through simulations, the proposed route discovery scheme using both EXIT and extended RREP message shows considerable' reduction in both route discovery time and control message overhead.

• 융합보안논문지
• /
• 제12권6호
• /
• pp.93-99
• /
• 2012

웹 애플리케이션 개발이 보편화됨에 따라 대부분의 소프트웨어는 개발기간 단축, 에러없는 품질, 수시 유지보수에 대한 적응성, 거대하고 복잡한 소프트웨어의 필요성 등과 같은 과제가 제기되었다. 웹 애플리케이션 개발시 이러한 품질문제에 대응하기 위해 소프트웨어의 재사용성, 신뢰성, 확장성, 단순성 등과 같은 측면을 고려하지 않을 수 없는 환경으로 변화 되었다. 이같은 상황에서 전통적 개발방법론으로는 보안품질을 해결하는데 한계를 가지고 있기 때문에 품질에 기반한 시큐어코딩 방법론이 필요하다. 웹 애플리케이션 보안품질은 애플리케이션 로직, 데이터, 아키텍처 전체 영역에서 별도의 방법론으로 대처하지 않으면 목표를 달성할 수 없다. 본 연구는 시큐어코딩의 최대 현안인 웹 애어플리케이션 개발을 위하여 웹 애플리케이션 아키텍쳐 설계절차를 제안한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권10호
• /
• pp.4618-4639
• /
• 2018

One of the most innovative paradigms for the next-generation of wireless cellular networks is the cloud-radio access networks (C-RANs). In C-RANs, base station functions are distributed between the remote radio heads (RHHs) and base band unit (BBU) pool, and a communication link is defined between them which is referred as the fronthaul. This leveraging link is expected to reduce the CAPEX (capital expenditure) and OPEX (operating expense) of envisioned cellular architectures as well as improves the spectral and energy efficiencies, provides the high scalability, and efficient mobility management capabilities. The fronthaul link carries the baseband signals between the RRHs and BBU pool using the digital radio over fiber (RoF) based common public radio interface (CPRI). CPRI based optical links imposed stringent synchronization, latency and throughput requirements on the fronthaul. As a result, fronthaul becomes a hinder in commercial deployments of C-RANs and is seen as one of a major bottleneck for backbone networks. The optimization of fronthaul is still a challenging issue and requires further exploration at industrial and academic levels. This paper comprehensively summarized the current challenges and requirements of fronthaul networks, and discusses the recently proposed system architectures, virtualization techniques, key transport technologies and compression schemes to carry the time-sensitive traffic in fronthaul networks.