• Journal of Information Processing Systems
• /
• 제10권4호
• /
• pp.491-502
• /
• 2014

In this paper we present some research results on computing intensive applications using modern high performance architectures and from the perspective of high computational needs. Computing intensive applications are an important family of applications in distributed computing domain. They have been object of study using different distributed computing paradigms and infrastructures. Such applications distinguish for their demanding needs for CPU computing, independently of the amount of data associated with the problem instance. Among computing intensive applications, there are applications based on simulations, aiming to maximize system resources for processing large computations for simulation. In this research work, we consider an application that simulates scheduling and resource allocation in a Grid computing system using Genetic Algorithms. In such application, a rather large number of simulations is needed to extract meaningful statistical results about the behavior of the simulation results. We study the performance of Oracle Grid Engine for such application running in a Cluster of high computing capacities. Several scenarios were generated to measure the response time and queuing time under different workloads and number of nodes in the cluster.

• Journal of Communications and Networks
• /
• 제14권5호
• /
• pp.554-565
• /
• 2012

In this paper, we study a cross-layer and end-to-end optimization problem for the integrated wireless and wireline network that consists of one wireline core network and multiple wireless access networks. We consider joint end-to-end flow control/distribution at the transport and network layers and opportunistic scheduling at the data link and physical layers. We formulate a single stochastic optimization problem and solve it by using a dual approach and a stochastic sub-gradient algorithm. The developed algorithm can be implemented in a distributed way, vertically among communication layers and horizontally among all entities in the network, clearly showing what should be done at each layer and each entity and what parameters should be exchanged between layers and between entities. Numerical results show that our cross-layer and end-to-end optimization approach provides more efficient resource allocation than the conventional layered and separated optimization approach.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2013년도 춘계학술발표대회
• /
• pp.144-147
• /
• 2013

클라우드 컴퓨팅은 낭비되는 컴퓨팅 자원들을 클라우드 서버를 통해 처리 가능한 고사양의 컴퓨팅 환경을 제공한다. 클라우드 컴퓨팅은 노드에 가상머신을 할당하여 사용자의 요구를 처리하는데, 노드의 공정한 사용과 사용자 요구사항 충족을 위해 FCFS, 라운드 로빈 등의 가상머신 할당 기법을 사용한다. 본 논문에서는 노드의 효율적인 사용을 위해 클라우드 서버에서 관리 중인 노드들의 성능을 미리 분석하고 가상머신을 할당 혹은 재할당시 분석된 결과를 바탕으로 할당할 노드를 선택하고 관리하는 시스템을 제안한다. 노드의 성능을 미리 분석함으로서 할당을 위한 가상머신 선택을 위한 연산을 줄여 신속하고 효율적인 가상머신 할당이 가능하다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2013년도 추계학술발표대회
• /
• pp.165-168
• /
• 2013

분산 클라우드 환경은 분산된 다수의 컴퓨팅 자원을 하나의 집합으로 묶고 하이퍼바이저를 이용하여 생성한 가상머신을 서비스 요구사항에 알맞게 할당하여 작업을 처리한다. 클라우드를 이용하여 제공하고자 하는 서비스는 이용되는 자원을 해당 서비스 대상 지역에 가장 높은 수준의 서비스 품질을 제공할 수 있는 데이터 센터로부터 할당 받는다. 그러나, 대부분의 클라우드 자원 제공자들은 해당 자원을 하나의 서비스 지역에 기반하여 할당한다. 따라서, 본 논문에서는 여러 지역에 애플리케이션 서비스를 제공하고자 하는 사업자들에게 전체 서비스의 최적 품질을 보장하는 다중 서비스 지역 기반 자원 할당 기법을 제안한다.

• 정보과학회 논문지
• /
• 제42권1호
• /
• pp.15-22
• /
• 2015

본 논문에서는 대규모의 계산 작업을 고성능으로 처리해야 하는 신약재창출 시뮬레이션 분야에 분산 슈퍼컴퓨팅 기술을 적용한 사례에 대해 논의하고자 한다. 신약재창출이란 기존에 알려진 약물의 새로운 적응증을 규명하는 것을 의미하며, 이러한 신약재창출은 비교적 짧은 수행시간을 갖는 대규모의 도킹(docking) 연산들을 고성능으로 처리해야한다는 점에서 Many-Task Computing (MTC) 성격을 지니고 있다. 이러한 MTC 응용들의 대표 사례로서 신약재창출 시뮬레이션을 분산 슈퍼컴퓨팅 환경 기반의 HTCaaS 시스템에 적용하였으며, 이를 통해 효율적인 작업 배포, 동적인 자원 할당 및 로드 밸런싱, 안정성 및 다양한 자원들의 효율적인 통합 등이 이러한 과학 응용들을 지원하는 데 있어 필수적인 기능임을 확인할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제5권9호
• /
• pp.1492-1512
• /
• 2011

Due to multiple hops, mobility and time-varying channel, supporting delay sensitive real-time traffic in wireless local area network-based (WLAN) mesh networks is a challenging task. In particular for real-time traffic subject to medium access control (MAC) layer control overhead, such as preamble, carrier sense waiting time and the random backoff period, the performance of real-time flows will be degraded greatly. In order to support real-time traffic, an efficient adaptive packet scheduling (APS) scheme is proposed, which aims to improve the system performance by guaranteeing inter-class, intra-class service differentiation and adaptively adjusting the packet length. APS classifies incoming packets by the IEEE 802.11e access class and then queued into a suitable buffer queue. APS employs strict priority service discipline for resource allocation among different service classes to achieve inter-class fairness. By estimating the received signal to interference plus noise ratio (SINR) per bit and current link condition, APS is able to calculate the optimized packet length with bi-dimensional markov MAC model to improve system performance. To achieve the fairness of intra-class, APS also takes maximum tolerable packet delay, transmission requests, and average allocation transmission into consideration to allocate transmission opportunity to the corresponding traffic. Detailed simulation results and comparison with IEEE 802.11e enhanced distributed channel access (EDCA) scheme show that the proposed APS scheme is able to effectively provide inter-class and intra-class differentiate services and improve QoS for real-time traffic in terms of throughput, end-to-end delay, packet loss rate and fairness.

• Journal of Information Processing Systems
• /
• 제2권2호
• /
• pp.88-94
• /
• 2006

In wireless cellular networks, previous researches on admission control policies and resource allocation algorithm considered the QoS (Quality of Service) in terms of CDP (Call Dropping Probability) and CBP (Call Blocking Probability). However, since the QoS was considered only within a predetermined cell capacity, the results indicated a serious overload problem of systems not guaranteeing both CDP and CBP constraints, especially in the hotspot cell. That is why a close interrelationship between CDP, CBP and cell capacity exists. Thus, it is indispensable to consider optimal cell capacity guaranteeing multiple QoS (CDP and CBP) at the time of initial cell planning for networks deployment. In this paper, we will suggest a distributed determination scheme of optimal cell capacity guaranteeing both CDP and CBP from a long-term perspective for initial cell planning. The cell-provisioning scheme is performed by using both the two-dimensional continuous-time Markov chain and an iterative method called the Gauss-Seidel method. Finally, numerical and simulation results will demonstrate that our scheme successfully determines an optimal cell capacity guaranteeing both CDP and CBP constraints for initial cell planning.

• Current Optics and Photonics
• /
• 제1권4호
• /
• pp.315-324
• /
• 2017

In recent years, multicast services such as high-definition television (HDTV), video conferencing, interactive distance learning, and distributed games have increased exponentially, and wavelength-division multiplexing (WDM) networks are considered to be a promising technology due to their support for multicast applications. Multicast survivability in WDM networks has been the focus of extensive attention since a single-link failure in an optical network may result in a massive loss of data. But the improvement of network survivability increases energy consumption due to more resource allocation for protection. In this paper, an energy-efficient multicast algorithm (EEMA) is proposed to reduce energy consumption in WDM networks. Two cost functions are defined based on the link state to determine both working and protection paths for a multicast request in WDM networks. To increase the number of sleeping links, the link cost function of the working path aims to integrate new working path into the links with more working paths. Sleeping links indicate the links in sleep mode, which do not have any working path. To increase bandwidth utilization by sharing spare capacity, the cost function of the protection path is defined to use sleeping fibers for establishing new protection paths. Finally, the performance of the proposed algorithm is evaluated in terms of energy consumption, and also the blocking probability is evaluated under various traffic environments through OPNET. Simulation results show that our algorithm reduces energy consumption while maintaining the quality of service.

• International Journal of Internet, Broadcasting and Communication
• /
• 제7권1호
• /
• pp.42-48
• /
• 2015

Generally, the current commercial cloud system is hosted by the centralization large datacenter. However, the other clouding forms existed and one of them is the P2P cloud. The P2P Cloud is a distributed systems, is freedom to join and leave and is possible to provide the IaaS. Commonly, P2P Cloud System is based on Pure-P2P System. Since a number of connection paths exist, it has a high reliability and easily scalability of the network. However, there are disadvantages of the slow rate of route because of looking up the all peers in order to search for the peer. In this paper, it proposes the P2P cloud system based on super-peer P2P. Super-peer P2P system has fast routing time, because of searching for cluster unit and it also can has high reliability and high scalability. We designs the super Peer cloud service and proposes the system model and Resource Allocation Algorithm for IaaS in Super peer P2P environment.

• 한국시뮬레이션학회논문지
• /
• 제19권2호
• /
• pp.81-89
• /
• 2010

생물학 데이터 마이닝은 생물학 데이터의 볼륨이 급격하게 증가함에 따라 최근 주목받고 있다. 그리드 기술은 계산 자원과 데이터 공유와 활용을 가능하게 한다. 이 논문에서는 생물학 데이터 마이닝과 그리드 기술을 결합한 혼합형 시스템을 제안한다. 특히, 생물학 데이터 마이닝의 처리 효율성을 위해 결정 범위 조정 알고리즘을 사용한다. 우리는 이 알고리즘을 통해 빠르고 자동으로 신뢰할 만한 데이터 마이닝 인식률을 얻는다. 게다가 그리드 환경에서는 지리적으로 분산된 자원들을 연동하기 때문에 통신량과 자원 할당이 이슈가 된다. 우리는 동적 로드 밸런싱을 제안하고 그리드 기반 생물학 데이터 마이닝 기법에 적용한 다. 성능 평가를 위해 우리는 평균 처리 시간, 평균 통신 시간, 평균 자원 활용도를 측정한다. 측정 실험의 결과는 제안된 두 알고리즘을 적용한 우리의 기법이 처리 시간과 비용 측면에서 이점을 제공한다는 것을 보여준다.