• Journal of the Korean Data and Information Science Society
• /
• v.10 no.2
• /
• pp.473-483
• /
• 1999

We introduce a way to construct a supercomputing capable system with some networked PCs, running the Linux operating system and computing power comparable with expensive commercial workstations, and with the Parallel Virtual Machine (PVM) software which enables one to control the total CPUs and memories of the networked PCs. By benchmarking the system using a PVM parallel program, we find that the system's parallel efficiency is close to 90 %.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.1
• /
• pp.67-75
• /
• 1997

This paper presents a virtual-parallel multistage interconnection network (MIN) which provides multipath between processor and memory module. The proposed virtual-parallel MIN network which uses $m{\times}1$ mutiplexer at the input switching block, $1{\times}m$ demultiplexer at the output switching block and logN-1 switching stages has maximum $2{\times}m$ unique paths between processor and memory module. Because it has multi-redundance paths, a number of processors can connect a specific Also, this new virtual-parallel structured MIN network can reduce packet collision possibility at switching block and it has cost. It shown to improve a performance and to be a very simple structure in comparision with MBSF structured MIN.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.41 no.6
• /
• pp.1-10
• /
• 2004

A haptic rendering system is proposed for a puncture task of a virtual vertebra model. To build a mesh model from medical images, Delaunay triangulation is applied and physical models are based on elasticity theory. Also, a redundant actuated 6 DOF parallel type haptic device is designed to display large force and to resolve the singularity problem of parallel type mechanisms. Haptic feeling of puncture task and the performance of the proposed haptic device are tested by two puncture task experiments.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.9
• /
• pp.189-196
• /
• 2020

Recently, GPU cloud computing technology applying GPU(Graphics Processing Unit) devices to virtual machines is widely used in the cloud environment. In a cloud environment, GPU devices assigned to virtual machines can perform operations faster than CPUs through massively parallel processing, which can provide many benefits when operating high-performance computing services in a variety of fields in a cloud environment. In a cloud environment, a GPU device can help improve the performance of a virtual machine, but the virtual machine scheduler, which is based on the CPU usage time of a virtual machine, does not take into account GPU device usage time, affecting the performance of other virtual machines. In this paper, we test and analyze the performance degradation of other virtual machines due to the virtual machine that performs GPGPU(General-Purpose computing on Graphics Processing Units) task in the direct path based GPU virtualization environment, which is often used when assigning GPUs to virtual machines in cloud environments. Then to solve this problem, we propose a GPGPU task management method for a virtual machine.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.74-74
• /
• 2016

복잡한 지형에서 컴퓨터를 이용한 물리적 기반 수치모의는 합리적인 시간내에 연산을 완료하기 위해 대개 큰 연산장비 들을 요구한다. 더욱이 모의되는 현상이 시간단계마다 갱신되어지는 동역학적 현상에 기반된 비정상상태일 때 연산성능은 고려되어지는 가장 중요한 주제가 될 수 있다. 연산 시간을 줄이기 위한 가장 널리 이용되는 전략중의 하나는 적절한 수의 프로세서를 이용하는 병렬 기법이다. 최근 들어 연산속도를 가속화하기 위해 다수의 코어를 이용한 OpenMP 와 MPI 기법들이 병렬해석기법으로 대두되었고 그래픽 연산장치를 이용한 병렬처리 해석기법도 소개되고 있다. 본 연구에서는 중앙연산장치를 이용한 병렬 해석기법을 이용하여 제내지 침수해석의 적용성을 검토하고 그 결과을 비교하였다. 본 연구를 위해 OpenMP 병렬기법을 이용하여 확산파 침수해석 프로그램의 원시코드를 재작성하여 가상 및 실제 유역에 적용하였다. 해석결과는 분산메모리 병렬해석 기법인 MPI를 도입한 모형의 결과와 비교되었다. OpenMP를 도입한 모형과 MPI를 도입한 경우 유량 및 수심의 경우 오차 허용 한계내에 수렴되어 만족되었으나 그러나 연산 속도의 경우 두 기법간의 자료의 저장 방법 차이로 인해 차이를 나타내었다. 가상 유역에 적용된 결과로 검토된 각 기법의 증속(speedup) 효과는 MPI의 경우 4 코어를 이용하였을 때 최고 2.62 배 정도에 도달하는 것으로 나타났다. OpenMP 를 적용한 경우 2.87 배 정도로 나타나 OpenMP 를 이용하였을 때 증속효과가 조금 더 뛰어났다. 이는 두 기법의 메모리 저장방식의 차이로 인해 자료의 전송량과 전송 시간이 적은 OpenMP 를 도입한 모형에서 MPI 모형 보다 상대적으로 뛰어난 결과를 나타내었다. 실제 유역의 적용을 위해 상대적으로 우수한 증속결과를 나타낸 OpenMP를 도입한 모형을 Malpasset 댐 붕괴 유역에 적용하였다. 적용된 요소의 수는 각각 45254, 11352 개로 비교적 많은 요소를 가진 하류지역에 적용하여 병렬효과를 극대화하고자 하였다. 적용결과 두 경우 모두 병렬 해석 기법을 도입한 모형에서 유속과 침수심 등은 순차적 모형과 동일한 값을 나타내었으나 증속효과로 인한 연산시간은 순차적 모형에서 8.57 배로 나타나 병렬 모형의 상대적으로 빠른 연산속도를 판단할 있었다. 위의 적용결과를 통해 계산 요소들이 많은 2 차원 해석의 경우 기존의 단일 코어를 이용한 순차적 해석은 장시간에 걸치 연산시간으로 인해 작업효율이 낮아지는 결과를 발생시킬 수 있으며 병렬 해석을 도입할 경우 주어진 컴퓨터 자원를 효율적으로 이용가능하여 합리적인 연산시간으로 연산결과를 얻는 것이 가능하여 반복적 통계 기법/Ensemble 해석 등을 이용한 종합적 해석이 좀 더 실용적으로 이루어 질 수 있을 것이라고 판단되었다.

• The KIPS Transactions:PartD
• /
• v.10D no.1
• /
• pp.117-124
• /
• 2003

The system implemented in this paper partitions and stores specimen data captured by a light microscope on distributed or parallel systems. Users ran observe images on computers as we use a physical microscope. Based on the client-server computing model, the system consists of client, coordinator, and data manager. Three components communicate messages. For retrieving images, we implemented the client program with necessary functions for educational applications such at image mark and text annotation, and defined the communication protocol. We performed the experiment for introducing a tape storage which stores a large volume of data. The experiment results showed performance improvement by data partitioning and indexing technique.

• Journal of KIISE
• /
• v.43 no.1
• /
• pp.27-33
• /
• 2016

Virtualization has become one of the most helpful techniques in computing systems, and today it is prevalent in several computing environments including desktops, data-centers, and enterprises. However, since I/O layers are implemented to be oblivious to the I/O behaviors on virtual machines (VM), there still exists an I/O scalability issue in virtualized systems. In particular, when a multi-queue solid state drive (SSD) is used as a secondary storage, each system reveals a semantic gap that degrades the overall performance of the VM. This is due to two key problems, accelerated lock contentions and the I/O parallelism issue. In this paper, we propose a novel approach, including the design of virtual CPU (vCPU)-dedicated queues and I/O threads, which efficiently distributes the lock contentions and addresses the parallelism issue of Virtio-blk-data-plane in virtualized environments. Our approach is based on the above principle, which allocates a dedicated queue and an I/O thread for each vCPU to reduce the semantic gap. Our experimental results with various I/O traces clearly show that our design improves the I/O operations per second (IOPS) in virtualized environments by up to 155% over existing QEMU-based systems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.4B
• /
• pp.604-612
• /
• 2000

In order to satisfy the desired QoS level associated with each existing connection, ATM networks require traffic policing during a connection. Users who respect the contract should receive the function of transparent traffic policing without any interruption. However, contract violations should be detected and mediated immediately. So we propose a CAM type policing controller to allow user cell streams to minimize additional delay. The proposed policing scheme controls policing actions including traffic shaping by suitably spacing cells on each virtual circuit. This policing action is based on parallel processing of multiple cell stream which arrive in ATM multiplexed virtual circuits. We have developed an analytical model of the proposed policing scheme to examine the amount of cell loss and delay, which depends on traffic load, the size of policing buffers and minimum spacing cell time.

• KIISE Transactions on Computing Practices
• /
• v.24 no.3
• /
• pp.129-137
• /
• 2018

NVMe(Non-Volatile Memory Express) SSD(Solid State Drive) is a high-performance storage that makes use of flash memory as a storage cell, PCIe as an interface and NVMe as a protocol on the interface. It supports multiple I/O queues which makes it feasible to process parallel-I/Os on multi-core environments and to provide higher bandwidth than SATA SSDs. Hence, NVMe SSD is considered as a next generation-storage for data-center and cloud computing system. However, in the virtualization system, the performance of NVMe SSD is not fully utilized due to the bottleneck of the software I/O stack. Especially, when it uses I/O stack of the hypervisor or the host operating system like Xen and KVM, I/O performance degrades seriously due to doubled-I/O stack between host and virtual machine. In this paper, we propose a new I/O engine, called Direct-AIO (Direct-Asynchronous I/O) engine, that can access NVMe SSD directly for I/O performance improvements on QEMU emulator. We develop our proposed I/O engine and analyze I/O performance differences between the existed I/O engine and Direct-AIO engine.

• Information and Communications Magazine
• /
• v.31 no.3
• /
• pp.39-45
• /
• 2014

본 고에서는 디지털 홀로그램의 생성 및 복원 방법에 관한 주요 이슈를 다룬다. 디지털 홀로그램 특히 컴퓨터 생성 홀로그램은 최근 디지털 디스플레이 및 고속 병렬 컴퓨팅 기술의 발달과 더불어 급속한 발전을 보이고 있다. 컴퓨터 생성 홀로그램은 메쉬, 점 집합 등으로 구성된 가상 객체의 표면에서 발생한 광파가 회절되는 분포를 수치 계산하여 생성된다. 이를 위하여 파동 방정식으로부터 유도되는 주요 회절식들을 살펴보고 이를 바탕으로 한 수치 계산 모델을 제시한다. 생성된 디지털 홀로그램을 광학 장치를 통한 재생 없이 수치적 복원을 통해 복원 결과를 확인하는 기법과 관련 이슈 또한 설명한다.