• Advances in aircraft and spacecraft science
• /
• 제4권3호
• /
• pp.297-316
• /
• 2017

With rapid growth in the complexity of large scale engineering systems, the application of multidisciplinary analysis and design optimization (MDO) in the engineering design process has garnered much attention. MDO addresses the challenge of integrating several different disciplines into the design process. Primary challenges of MDO include computational expense and poor scalability. The introduction of a distributed, collaborative computational environment results in better utilization of available computational resources, reducing the time to solution, and enhancing scalability. SORCER, a Java-based network-centric computing platform, enables analyses and design studies in a distributed collaborative computing environment. Two different optimization algorithms widely used in multidisciplinary engineering design-VTDIRECT95 and QNSTOP-are implemented on a SORCER grid. VTDIRECT95, a Fortran 95 implementation of D. R. Jones' algorithm DIRECT, is a highly parallelizable derivative-free deterministic global optimization algorithm. QNSTOP is a parallel quasi-Newton algorithm for stochastic optimization problems. The purpose of integrating VTDIRECT95 and QNSTOP into the SORCER framework is to provide load balancing among computational resources, resulting in a dynamically scalable process. Further, the federated computing paradigm implemented by SORCER manages distributed services in real time, thereby significantly speeding up the design process. Part 1 covers SORCER and the algorithms, Part 2 presents results for aircraft panel design with curvilinear stiffeners.

• Advances in aircraft and spacecraft science
• /
• 제4권3호
• /
• pp.317-334
• /
• 2017

With rapid growth in the complexity of large scale engineering systems, the application of multidisciplinary analysis and design optimization (MDO) in the engineering design process has garnered much attention. MDO addresses the challenge of integrating several different disciplines into the design process. Primary challenges of MDO include computational expense and poor scalability. The introduction of a distributed, collaborative computational environment results in better utilization of available computational resources, reducing the time to solution, and enhancing scalability. SORCER, a Java-based network-centric computing platform, enables analyses and design studies in a distributed collaborative computing environment. Two different optimization algorithms widely used in multidisciplinary engineering design-VTDIRECT95 and QNSTOP-are implemented on a SORCER grid. VTDIRECT95, a Fortran 95 implementation of D. R. Jones' algorithm DIRECT, is a highly parallelizable derivative-free deterministic global optimization algorithm. QNSTOP is a parallel quasi-Newton algorithm for stochastic optimization problems. The purpose of integrating VTDIRECT95 and QNSTOP into the SORCER framework is to provide load balancing among computational resources, resulting in a dynamically scalable process. Further, the federated computing paradigm implemented by SORCER manages distributed services in real time, thereby significantly speeding up the design process. Part 1 covers SORCER and the algorithms, Part 2 presents results for aircraft panel design with curvilinear stiffeners.

• 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.

• 인터넷정보학회논문지
• /
• 제17권5호
• /
• pp.67-74
• /
• 2016

Distributed agile software development faces difficulties for instance lack of visibility across development and delivery processes, complex and disjointed development processes, inability to capitalize on business opportunities, lack of communication agility between disconnected owners, development teams, and users or clients. However these difficulties are solved with the help of cloud computing services. This study proposes a framework to provide a skeletal or structural environment for distributed agile software development in cloud computing environment. The framework guide towards the best tooling to deliver a consistent, automated, governed, and unified agile software development process with reduced technical debt, and minimized project backlog. In addition to this, the study highlights the benefits of cloud computing in agile software development.

• 한국데이타베이스학회:학술대회논문집
• /
• 한국데이타베이스학회 1994년도 DB산업기술 활성화를 위한 학술대회 및 기술 심포지움
• /
• pp.105-129
• /
• 1994

The information system of corporations are facing a new environment expressed by miniaturization, decentralization and Open System. It is therefore of utmost importance for corporations to adapt flexibly th such new environment by providing for corresponding changes to their existing information systems. The objectives of this study are to identify this new environment faced by today′s information system and develop effective methods for data resource management under this new environment. In this study, it is assumed that the new environment faced by information systems can be specified as Distributed Computing Environment, and in order to achieve such system, presents Client/server architecture as its representative computing structure, This study defines Client/server architecture as a computing architecture which specialize the fuctionality of the client system and the server system in order to have an application distribute and perform cooperative processing at the best platform. Furthermore, from among the five structures utilized in Client/server architecture for distribution and cooperative processing of application between server and client this study presents two different data management methods under the Client/server environment; one is "Remote Data Management Method" which uses file server or database server and. the other is "Distributed Data Management Method" using distributed database management system. The result of this study leads to the conclusion that in the client/server environment although distributed application is assumed, the data could become centralized (in the case of file server or database server) or decentralized (in the case of distributed database system) and the data management method through a distributed database system where complete responsibility and powers with respect to control of data used by the user are given not only is it more adaptable to modern flexible corporate environment, but in terms of system operation, it presents a more efficient data management alternative compared to existing data management methods in terms of cutting costs.

• 한국전자통신학회논문지
• /
• 제1권1호
• /
• pp.20-26
• /
• 2006

최근에 급성장하고 있는 실시간 통신 분산 컴퓨팅은 최근에 컴퓨터 응용분야의 하나로서 컴퓨터 과학과 공학 분야에서 급성장하고 있는 한 분야이다. 실시간 객체 지향 분산 컴퓨팅은 분산된 컴퓨터 시스템에서 객체 네트워크의 형태로 구성된 실시간 분산 컴퓨팅이다. 본 논문에서는 TMO 구조를 이용하여 실시간 통신 시뮬레이션 프로그래밍을 하기 위해 DHS(Distributed High-Precision Simulation)라는 응용 환경에 적용하였다. TMO 구조에서 사용되는 시간 구동 및 메시지 구동 구조는 충분한 서비스를 제공하기 위해 데드라인 시간을 보장함으로서 실시간 시뮬레이션 프로그래머의 노력을 줄일 수 있는 이점을 가지고 있다. 데드라인 시간 보장은 프로그래머의 설계 시간을 처음 객체부터 적용하여 보장함으로서 형성될 수 있다. 분산된 객체지향 실시간 통신 시스템에서 TMO 구조를 이용한 다단계 프로그래밍 설계를 함으로서 실시간 통신 프로그래밍을 유연하게 작성할 수 있다.

• 인터넷정보학회논문지
• /
• 제7권6호
• /
• pp.87-103
• /
• 2006

컴퓨터와 네트워크 기술의 향상은 예전에는 슈퍼컴퓨터에서나 가능한 일을 분산 처리할 수 있는 환경적 기반을 제공한다. 분산 컴퓨팅 환경을 제공하기 위해서는 우선 분산 런타임 시스템이 구축되어야 하는데 기존의 전통적인 분산 런타임 시스템들은 대부분이 정적인 마스터 노드와 작업 노드들로 구성되는 구조를 갖기 때문에 분산처리 작업량의 변동에 따라 시스템을 유연하게 동적으로 재구성할 수 없다는 단점을 갖는다. 이에 본 논문에서는 P2P 환경에서의 분산 런타임 시스템인 작업 할당 관리자의 모델을 제시하고 구현하여 유연하고 동적인 시스템 구축이 가능하도록 하였다. 즉, P2P 표준 프로토콜인 JXTA 플랫폼 상에서 협업 환경을 위해 개발자들 간에 작업 프로그램의 전달과 관리, 그리고 원격 컴파일과 실행 작업들을 수행할 수 있도록 하였다 이 방식은 유연하고 동적인 시스템 구축이 가능하기 때문에 작업의 분산처리를 위해 필요한 유휴 자원들을 필요한 시점에 즉시 확보하여 활용할 수 있다는 장점을 가진다. 이와 더불어 인터넷 정보검색을 위해 방대한 데이터를 수집하는 크롤러를 본 논문에서 구현한 시스템을 이용하여 분산 처리시킴으로써 본 시스템의 유용성과 분산처리 성능을 보여 줄 수 있도록 하였다.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2000년도 추계종합학술대회
• /
• pp.368-371
• /
• 2000

분산 컴퓨팅 기술과 객체지향 기술의 발전은 인터넷상의 동일 기종간의 분산처리뿐만 아니라, 이 기종간에도 분산 컴퓨팅이 가능한 분산 객체 컴퓨팅으로 발전되었다. 이를 대표하는 CORBA(Comm on Object Request Broker Architecture)도 단지 네트워크를 통한 이기종간의 상호 운용성만을 해결한 상태이며, 대부분 서버 객체의 존재 및 정상적인 활동이 보장되지만 네트워크 오버헤드가 발생하는 경우에는 클라이언트의 서비스 요구에 응할 수 없는 문제점을 가지고 있다. 이러한 문제를 해결하는 방안 중 하나로, 데이터와 코드를 결합한 객체를 직접 이동하는 이동에이전트(Mobile Agent)를 분산 컴퓨팅 기술과 결합하는 방법이 대두되고 있다.

• 한국멀티미디어학회논문지
• /
• 제19권8호
• /
• pp.1487-1497
• /
• 2016

Explosively growing service of digital multimedia data increases the need for highly scalable low-cost storage. This paper proposes the new storage architecture based on torus network which does not need network switch and erasure coding for efficient storage usage for high scalability and efficient disk utilization. The proposed model has to compensate for the disadvantage of long network latency and network processing overhead of torus network. The proposed storage model was compared to two most popular distributed file system, GlusterFS and Ceph distributed file systems through a prototype implementation. The performance of prototype system shows outstanding results than erasure coding policy of two file systems and mostly even better results than replication policy of them.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2015년도 추계학술발표대회
• /
• pp.227-229
• /
• 2015

모든 사물에 인터넷이 연결되는 사물 인터넷(IoT: Internet of Things)시대가 열렸다. IoT 디바이스들을 연결하기 위해 클라우드 또한 더욱 관심이 높아지고 있다. IoT 디바이스를 연결한 클라우드는 작은 단위의 작업들을 다량으로 수행하게 된다. IoT 서비스에서 발생하는 작업들을 효율적으로 처리하기 위해서는 적합한 작업 스케줄링이 반드시 필요하다. 본 논문에서는 오픈소스 기반의 플랫폼인 오픈스택(OpenStack)에서 Filter 스케줄러와 Chance 스케줄러를 VM 개수에 따라 단위 시간동안 성능을 비교 분석한다. 이를 통해 오픈스택에서 IoT 서비스 사용자들을 위해 합리적인 스케줄러 방법을 도출해낼 수 있다