• The Magazine of the IEIE
• /
• v.22 no.9
• /
• pp.14-27
• /
• 1995

KAICUBE/한빛-1호는 하이퍼큐브 형태의 연결망을 가진 병렬 컴퓨터이고 각 노드는 i860프로세서와 통신용의 i82380 DMA 콘트롤러를 탑재하고 있다. 40Mh2 CPU클럭을 사용하는 32노드로 구성되어 있고 컴퓨터의 최고 속도는 2.5G-flops 정도로써 이것은 국내 최초의 Giga급 컴퓨터이다. DMA콘트롤러에 의해 구동되는 노드간 통신은 채널 대역폭이 100Mbps정도이다. 0번 노드는 UNIX를 탑재한 호스트 컴퓨터와 연결되어 있고 호스트 컴퓨터는 병렬 프로그래밍 환경과 각 노드를 관리하는 역할을 한다. 익스프레스는 호스트 컴퓨터에 탑재된 병렬 운영 체제이고 사용하기 간편한 사용자 환경과 프로그래밍 방법에 따라 호스트-노드방법과 cubits 프로그래밍 환경을 각각 제공한다. 그밖에 고수준의 병렬 프로그래밍 환경으로써 기존의 순차 프로그램에 기초한 입력 프로그램을 병렬 프로그램으로 자동 변환 해주는 KAPPA가 있다. 여러 분야의 과학 계산용 프로그램이 수행되고 있으며 그의 성능 측정을 통하여 탁월한 성능을 보여 주었다. 보다 편리한 병렬 프로그래밍 환경의 개발과 범용 계산 전응 서버로써 자유로이 사용할 수 있도록 네트워크 기능을 강화하는 일이 남아있다.

• The Journal of the Korea Contents Association
• /
• v.2 no.2
• /
• pp.103-111
• /
• 2002

In parallel application programs with a localized communication, even if the MINs have lour diameters, overall system performance degrades when compared to the hypercube and tree structure. The reason is that it is impossible for MINs to provide some mechanisms for clustering to exploit the locality of reference. However proposed MIN can be constructed suitable for localized communication by providing the shortcut path and multiple paths inside the processor-memory duster which has frequent data communications. Therefore proposed MIN achieves enhanced performance in parallel application program with a localized communication.

• Proceedings of the IEEK Conference
• /
• 1998.06a
• /
• pp.293-296
• /
• 1998

In this paper, we propose a matrix hypercube graph as a new topology for parallel computer and analyze its characteristics of the network parameters, such as degree, routing and diameter. N-dimensional matrix hypercube graph MH(2,n) contains 22n vertices and has relatively lower degree and smaller diameter than well-known hypercube graph. The matrix hypercube graph MH(2,n) and the hypercube graph Q2n have the same number of vertices. In terms of the network cost, defined as the product of the degree and diameter, the former has n2 while the latter has 4n2. In other words, it means that matrix hypercube graph MH(2,n) is better than hypercube graph Q2n with respect to the network cost.

• The Magazine of the IEIE
• /
• v.20 no.7
• /
• pp.104-114
• /
• 1993

본 연구 개발에서는 계속적으로 그 필요가 증가하는 대규모 계산을 위한 슈퍼급의 성능을 가지는 병렬처리 시스템의 구현, 구축을 목표로 하고 있다. 우리는 슈퍼급의 컴퓨터 시스템을 개발하기 위해 최신의 i860 프로세서를 사용한 단위컴퓨터를 설계, 제작하여 최대 40 MFLOPB의 성능을 가지는 단위 컴퓨터를 구현하였고, 8개의 단위컴퓨터를 연결하여 320MFLOPS컴퓨터를 개발하였다. 최종 목표는 128개의 단위 컴퓨터를 연결하여 5 GFLOPS 최대 계산능력을 갖는 7차원 하이퍼큐브 컴퓨터 시스템 KAICUBE-860을 구성하고자 한다. 대부분의 사람들은 그 생각방식이나 행동형태가 순차적인 구조를 가지고 있기 때문에 병렬적인 프로그램의 개발이 쉽지가 않고 때문에 사용자로 하여금 보다 쉽게 병렬시스템에 접근할 수 있도록 사용자 환경을 적절히 구축하여야만 한다. 따라서 우리의 최종 목표는 슈퍼급 성능의 병렬 처리 시스템이고 그 사용 방법은 보통순차적 컴퓨터를 사용하듯이 간단한 사용자 환경을 제공하는 시스템이다. 이를 위해 시스템 소프트웨어와 사용자 환경을 계속 개발 중에 있다.

• 한국전산유체공학회:학술대회논문집
• /
• 2008.03a
• /
• pp.229-232
• /
• 2008

This study presents a numerical procedure to optimize the shape of stepped circular pin-fins to enhance turbulent heat transfer. The KRG method is used as an optimization technique with Reynolds-averaged Navier-Stokes analysis of fluid flow and heat transfer with shear stress transport turbulent model. The objective function is defined as a linear combination of heat transfer and friction loss related terms with a weighting factor. Ten training points are obtained by Latin Hypercube Sampling for two design variables. Optimum shape has been successfully obtained with the increased objective function.

• 한국전산유체공학회:학술대회논문집
• /
• 2008.10a
• /
• pp.229-232
• /
• 2008

This study presents a numerical procedure to optimize the shape of stepped circular pin-fins to enhance turbulent heat transfer. The KRG method is used as an optimization technique with Reynolds-averaged Navier-Stokes analysis of fluid flow and heat transfer with shear stress transport turbulent model. The objective function is defined as a linear combination of heat transfer and friction loss related terms with a weighting factor. Ten training points are obtained by Latin Hypercube Sampling for two design variables. Optimum shape has been successfully obtained with the increased objective function.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.7
• /
• pp.952-965
• /
• 1995

In the hypercube multiprocessor system. repeated allocations and deallocations of subcubes generate a fragmented hypercube from which. even if sufficient free processors are available, a subcube that is large enough to accomodate a new task cannot be formed. To eliminate the fragmentation, we need a processor relocation to move a task-occupied subcube to a free subcube, and gather dispersed small subcubes. In this paper, we propose a routing algorithm for task migration in circuit- switched hypercube to relocate processors. In the circuit-switched hypercube, we have to find a set of dedicated link-disjoint routing paths for each node to move a task from a busy subcube to a free subcube in fragmentation. The proposed algorithm is based on the PGC(packed Gray code) which detects all kinds of subcubes.

• Proceedings of the IEEK Conference
• /
• 2003.11b
• /
• pp.11-14
• /
• 2003

In interconnection network for parallel processing, the cycle partitioning problem for parallel transmission with faulty vertieces or edges is very important. In this paper, we assume that k($\leq$m-1) edges do not share any vertices of m dimension hypercube Q$_{m}$ and show that it is possible to construct a cycle set which consists of k cycles covering all the vertices of the hypercube and one cycle including one of the given edges. This cycle set can be used to parallel transmission between two vertices joined by faulty edges.s.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.633-636
• /
• 2000

본 논문에서는 하이퍼큐브보다 망비용이 개선된 HCN(n,n)과 2n-hypercube 사이의 임베딩을 분석한다. 2n-hypercube를 HCN(n.n)에 연장율 3에 임베딩 가능함을 보이고, HCN(n,n)을 2n-hypercube에 임베딩하는 비용이 O(n)임을 보인다.

• The KSFM Journal of Fluid Machinery
• /
• v.11 no.1
• /
• pp.52-60
• /
• 2008

Staggered dimples printed on opposite walls of an internal cooling channel are formulated numerically and optimized to enhance heat transfer performance. Nusselt number and friction factor based objectives are considered and a weighted average surrogate model is used to approximate the data generated by numerical simulation. The dimpled channel shape is defined by three geometric design variables, and the design point within design space are selected using Latin hypercube sampling. A weighted-sum method for multi-objective optimization is applied to integrate multiple objectives into a single objective. By the optimization, the objective function value is improved largely and heat transfer rate is increase much higher than pressure loss increase due to shape deformation. Channel with vertically non-symmetric optimum dimples is tested and found that the best appears if dimples on opposite wall are displaced by one quarter of dimple spacing.