• Journal of the Korea Society for Simulation
• /
• v.7 no.1
• /
• pp.69-83
• /
• 1998

In this paper, we present the results of simulation by running parallel VHDL simulation on typical MPP(Massively Parallel Processor) systems such as IBM SP2 and SGI Origin 2000. Parallel simulation uses the synchronous protocol and parallel program is implemented using MPI(Message Passing Interface) based on message passing model, so that it can urn on any parallel programming environment which supports MPI, a standard communication library. And then GVT(Global Virtual Time) computation for parallel simulation is based on the global broadcasting with MPI＿Bcast(), which is a standard function in MPI and piggybacking. Our benchmark exhibits that as size of VHDL grows, the parallel simulation has a better performance compared with the sequential simulation. In addition, we also show the results of comparison between IBM SP2 and SGI Origin 2000 by applying the same application to those indirectly.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.5 s.37
• /
• pp.49-56
• /
• 2005

Reuse of software is the process of developing software system which does not develop all modules newly but construct system by using existing software modules. Object oriented concept introduces dynamic binding and polymorphism, which increases number of executable Paths. In this case static analysis of source code is not useful to find executable path. These concepts occur new Problems in retesting. Tradition testing method can not cover complex relations such as inheritance, message Passing and instance of and can not apply on object oriented retesting. we propose the method to reduce the number of methods and classes to be retested in associated class relation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.8
• /
• pp.637-642
• /
• 2014

In this paper, we propose a efficient shift index searching algorithm for design of the block LDPC codes. It is combined with the message-passing based cycle search algorithm and ACE algorithm. We can determine the shift indices by ordering of priority factors which are effect on the LDPC code performance. Using this algorithm, we can construct the LDPC codes with low complexity compare to trellis-based search algorithm and save the memory for storing the parity check matrix.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.5
• /
• pp.450-463
• /
• 2000

Array redistribution is usually required to enhance algorithm performance in many parallel programs on distributed memory multicomputers. GEN_BLOCK redistribution, which is redistribution between different GEN_BLOCKs, is essential for load balancing. However, prior research on redistribution has been focused on regular redistribution, such as redistribution between different CYCLIC(N)s. GEN_BLOCK redistribution is very different from regular redistribution. Message passing in regular redistribution involves repetitions of basic message passing patterns, while message passing for GEN_BLOCK redistribution shows locality. This paper proves that two optimal condition, reducing the number of communication steps and minimizing redistribution size, are essential in GEN_BLOCK redistribution. Additionally, by adding a relocation phase to list scheduling, we make an optimal scheduling algorithm for GEN_BLOCK redistribution. To evaluate the performance of the algorithm, we have performed experiments on a CRAY T3E. According to the experiments, it was proven that the scheduling algorithm shows better performance and that the conditions are critical in enhancing the communication speed of GEN_BLOCK redistribution.

• Journal of Korea Society of Industrial Information Systems
• /
• v.14 no.5
• /
• pp.19-32
• /
• 2009

In this paper, we design thin-client framework capable of application sharing & data access on the Internet, and apply related skills, such as X windows system, pseudo server, CODA file system, MPI(Message Passing Interface). We suggest a framework for the thin client to access data produced by working on a server optimally as well as to run server side application, even in the case of network down. Additionally, it needed to reflect all local computing changes to remote server when network is restored. To design thin client framework with these characteristics, in this paper, we apply distributed pseudo server and CODA file system to our framework, also utilize MPI for the purpose of more efficient computing & management. It allows for implementation of network independent computing environment of thin client, also provide scalable application service to numerous user through the elimination of bottleneck on caused by server overload. In this paper, we discuss the implementing method of thin client framework in detail.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.39 no.2
• /
• pp.191-197
• /
• 2015

The message passing interface (MPI) and hybrid programming models for the parallel computation of a pressure equation were compared in a distributed memory system. Both models were based on domain decomposition, and two numbers of the sub-domain were selected by considering the efficiency of the hybrid model. The parallel performances for various problem sizes were measured using up to 96 threads. It was found that in addition to the cache-memory size, the overhead of the MPI communication/OpenMP directives affected the parallel performance. For small problems, the parallel performance was low because the percentage of the overhead of the MPI communication/OpenMP directives increased as the number of threads increased, and MPI was better than the hybrid model because it had a smaller communication overhead. For large problems, the parallel performance was high because, in addition to the cache effect, the percentage of the communication overhead was relatively low compared to that for small problems, and the hybrid model was better than MPI because the communication overhead of MPI was more dominant than that of the OpenMP directives in the hybrid model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.4
• /
• pp.637-640
• /
• 2022

To maximize performance in a high-performance multicore processor system. it is essential to enable effective data communication between processing cores. Data communication between processor nodes can be broadly classified into collective and point-to-point communications. Collective communication comprises scattering and gathering. This paper presents a efficient message scattering and gathering based on processing node status. In the proposed algorithms, the transmission order is changed according to the data size of the pre-existing communication, to reduce the waiting time required until the collective communications begin. From the simulation, the performances of the proposed message scattering and gathering algorithms were improved by approximately 71.41% and 69.84%.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.8
• /
• pp.2158-2170
• /
• 1999

It has been required more reliable communication on processes and improvement of system performance as distributed systems using multicast protocol became widespread. In distributed environment maintaining data consistency through asynchronous execution of processes and coordinating the activities of them would occurs. This paper proposes a total ordering algorithm, TORMP, in order to resolve these problems. TORMP takes advantage of multicast protocol and uses an effective token passing method. It reduces a process delaying time before transmitting its message by multicasting a token simultaneously to every process that initiates the request of the message. Moreover, the processes receiving the token start multicasting the message at the same time, which causes to cut down the overall transmission dely. In case that one process sends a message, TORMP hardly uses the procedure of controlling for ordering. It gives fairly the right of sending messages to all processes in a group with utilizing vector clock. In TORMP, unlike other algorithms, the number of packets generated during ordering process does not depend on the number of processes.

• 한국전산유체공학회:학술대회논문집
• /
• 2006.05a
• /
• pp.408-411
• /
• 2006

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.4
• /
• pp.926-938
• /
• 1996

In this paper, we proposed a new multicast algorithm which is efficiently applicable to overlapped process group environments where one process may be involved in several process groups. Unlike the esisting algorithms, the proposed one provides an efficient group communication mechanism by generating the processgroup information in two-level tree. Using this algorithm, we improved the shortcoming of the existing algorithms by reducing the overhead in passing through unnecessary processes for message ordering. We have provided the causal ordering method as well as the total ordering method in group communication environments. As a result, we allow one process to deliver message to other processes with a short delay time, and reduced the overhead required for the message ordering. Also, we logically proved the proposed causal ordering method, and compared the performance of the proposed algorithm with ones of other existing algorithms by computer simulation.