Earticle

본 연구에서는 생산 자료의 역산을 통해 투과도, 상대투과도와 대수층 물 유입량을 규명할 수 있는 메모리분산 병렬처리 GA에 기반한 역산 모델을 개발하였다. 이 모델에서 상대투과도는 코아 분석 등을 통해 선정된 상대투과도 모델 중에서 최적의 모델을 GA가 자동으로 선택하도록 개발되었다. 개발된 모델을 3개 층으로 구성된 불균질 저류층에 적용하여 3년간 관측된 생산 자료의 역산을 통해 투과도, 상대투과도, 물 유입량을 규명하는 작업을 PC-클러스터에서 수행하였다. 역산을 수행한 결과, 30세대만에 수렴되었으며 이 때 적합도 값이 0.000112이였다. 생산 압력 매칭 결과, 생산정 5에서 최대 상대오차가 3.1%로서 산출된 상대투과도, 투과도, 물 유입량은 신뢰할 만한 결과임을 알 수 있다. 상대투과도 모델 수렴양상을 살펴보면, 1세대에서 20개씩 동일했던 개체들은 5세대만에 한 개의 상대투과도 모델로 수렴되었다.

This paper presents the development of inverse model equipped with the memory-shared parallel genetic algorithm (GA) to characterize permeability, water influx and relative permeability, simultaneously. The optimum relative permeability model is selected by GA automatically among the relative permeability models obtained by core analysis in the developed model. By utilizing that, we have estimated the permeability distribution and water influx, and selected the relative permeability model for 3-layered heterogeneous reservoir system with the observed production data during three years. This work was carried out with PC-cluster. From the results of the inverse calculation, it was found that it was converged at thirty generations, and the fitness of the best individual was 0.000112. The pressure matching results showed that the maximum relative error was 3.1% at the fifth producer, and the developed model has generated the favorable permeability distributions, water influx and relative permeability model. The process of selecting the relative permeability model was as follows; The number of individual in each node was twenty at the first generation, but it was converged into the only relative permeability model at fifth generation.

최근의 컴퓨터의 응용영역은 클라우드 컴퓨팅, 사물인터넷, 인메모리 컴퓨팅, 빅데이터 등으로 변화하고 있으며, 이에 따라 컴퓨터가 필요로 하는 메모리의 양은 기하급수적으로 증가하고 있다. 또한 고성능 컴퓨팅, 수퍼컴퓨팅 등공유 메모리를 필요로 하는 응용영역 역시 확대되고 있는 추세이다. 기존의 중앙처리장치(CPU)에서는 칩 면적의한계로 인하여 더 이상 메모리를 증설할 수 있는 공간이 존재하지 않으므로 메모리의 확장이 어렵고, 기존방식에서처럼 I/O를 이용한 분산공유메모리는 억세스하기에 필요한 지연시간이 매우 길기 때문에 최근의 복잡한 계산과 데이터의 이동성을 중요시하는 응용영역에서는 적합하지 않다. 본 논문에서는 인터커넥션의 개수를 최소화함으로써중앙처리장치의 칩 면적을 많이 차지하지 않으면서도 성능면에서는 I/O를 이용한 방식보다 낮은 지연시간의 효과를가져올 수 있는 직렬 메모리 접속기술을 소개하며, 이를 위한 메모리 컨트롤러 구조를 제시하고, 또한 실제 임베디드 프로세서를 이용한 시스템에 이를 적용한 결과를 보인다. 또한 본 논문에서 제시하는 직렬 메모리 접속기술을 이용하여 공유메모리를 구성하는 방법을 제안한다.

Recently the computer's application areas are changing to new realm such as cloud computing, internet of things, in-memory computing, big data and etc. and thus the amount of memory that computer needs is increased exponentially. In addition, there is a trend that the application areas that need shared memory is expanding, for example, high-performance computing, supercomputing, and so on. Since in a conventional Central Processing Unit (CPU), due to limitations of chip area, it does not have any more space to install more memory. So it is difficult to expand the memory and distributed shared memory using the I/O as in the conventional method. And it is not appropriate in the application areas that have importance at recent complex calculation and data mobility because the delay time is too long. This paper introduces the serial memory access technique that can bring the effect of the lower delay than using I/O, not taking up too much chip area of the CPU by minimizing the number of interconnections, propose the structure of the memory controller for it, and show the result of applying it to the system using actual embedded processor system. In addition, we propose a method for configuring share memory by using the serial memory access technique presented in this paper.

이기종 처리환경은 이종의 CPU/GPU가 혼재하는 다처리 환경으로서 빠른 연산속도를 필요로 하는 다양한 응용 분야에서 리 사용되고 있다. 이러한 이기종 처리환경은 CPU/GPU/APU등 로세서를 통합리하여 데이터 공 유가 가능하며 력, 소형화, 고성능을 지향하는 계산 환경변화에 따라 이기종 처리 시스템은 지속으로 발하 고 있다. 여러 가지 응용분야 상처리는 이기종 환경에 가장 합하다. 필터링, 보간, 히스토그램 계산과 같은 상처리 련 연산은 공유 지역 메모리를 효율으로 사용할 수 있는 연산이므로 특별히 이기종 처리 시스템에서 효 과가 극화 된다. 본 논문에서는 이기종 처리 시스템에서의 공유 메모리가 가지는 효과를 소개하고 공유 메모리를 사용하는 상처리 기능을 통해 캐쉬 공유가 효율임을 보이고 이를 통해 CPU와 GPU간의 효율인 작업을 분배 방법을 제안한다.

Heterogeneous processing systems are multi-processing environment containing heterogeneous CPU and GPU. They are widely used in a variety of areas that require fast computation speed. Recently, it is becoming possible to share address space and data through the integration of GPU and CPU. Furthermore, on-die integration of GPU architecture is developing as low power consumption, miniaturization and high performance are becoming more important. Operations in image processing such as filtering, interpolation, and histogram computation can efficiently use shared local memory and applying it to the on-die integration of GPU system in which CPU and GPU share the memory and address system can maximize the effect of those systems. We introduced the study of the effects of shared memory in the existing on-die integration of GPU system, we propose a method of efficient workload distribution between CPU and GPU by measuring the efficiency of cache sharing through image processing operations that use the shared local memory space.

팀 조직은 기업의 대규화가 시작된 1950년대에 제안된 이후 조직운영의 기본형태가 되어가고 있다. 급격하게 변화하는 기업환경에서 조직의 수행직무는 전문적이고 창조적 능력, 복합적인 기술력, 정확한 판단력과 질적 경험이 필요하다. 그리고 팀 구성원들의 상호작용성과 상호의존성이 팀의 성공여부를 좌우함으로, 이를 위해 조직운영의 새로운 그 무엇인가가 요구되어진다. 이와 같은 맥락에서 인지론적 관점에서 이슈가 되고 있는 공유멘탈모델, 정보교류체제 및 팀 효능감 등의 요인이 팀의 수행과 어떠한 관계가 있는지를 실증적으로 분석하고자 한다. 이러한 연구결과를 반영하여 효율적이고 성공적인 팀제운영에 일조할 것으로 예측되어진다. 본 연구목적을 달성하기 위하여 이론적 배경을 기초로 연구모형과 연구가설을 설정하고 검정한 결과는 다음과 같다. 첫째, 공유멘탈모델의 과업모델은 정보교류체제의 전문성, 신뢰성 및 과업조정에 유의한 영향요인이며, 팀 모델은 정보교류체제의 신뢰성과 과업조정요인에 유의한 영향을 미치는 것으로 분석되었다. 둘째, 공유멘탈모델의 과업모델과 팀 모델 모두는 팀 성과에 긍정적인 영향을 미치지 않으며, 정보교류체제의 신뢰성과 과업조정요인은 팀 성과에 영향을 미치는 것으로 분석되었다. 셋째, 정보교류체제의 전문성과 과업조정 요인이 팀 효능감에 긍정적인 영향을 미치며, 팀 효능감은 팀 성과에 유의한 영향을 미치고 있다. 넷째, 정보교류체제의 전문성과 과업조정 요인과 팀 성과사이에서 팀효능감이 매개역할을 수행하는 것으로 분석되었다. 이상과 같은 연구결과는 기업의 경쟁 환경이 치열해질수록 생존과 성장을 위하여 구성원간 상호작용성과 의존도가 커질 것이라는 점을 고려한다면 향후 조직운영의 효율성을 강화에 시사하는 바가 크다고 볼 수 있다. 또한, 향후 이를 보완하는 다양한 변수를 고려하여 연구를 진행한다면 조직운영에 큰 도움을 줄 수 있을 것이다.

The purpose of this study empirically identify the relationship of shared mental model(SMM), transactive memory system(TMS), team efficacy, and team performance. In order to achieve the objectives of this paper, the study was conducted in parallel the theoretical studies through the preceding literature review as well as the empirical test through questionaries analysis. First, in the relationship between SMM and TMS, task model of SMM is significantly associated with all three dimensions of TMS(specialization, credibility, and task coordination). And, team model of SMM is significantly associated with two dimensions of TMS( credibility and task coordination). The fact that the team model of SMM do not affect the specialization factor of TMS means that the specialization factor already know who have any expertise among the team members. This is attributed to the homogeneity of team model of SMM and specialization of TMS. Second, in the relationship between SMM and team performance, both task model and team model showed that there is not a positive impact on team performance. The results are contrary to the results of previous studies. These results is because the study was mainly conducted with public organizations and organizations diverted from public organization in the past to private organizations. It is inferred that these organizations relatively weak the need for shared mental and are still adopting the introduction and operation of team system, such as the remaining of vertical hierarchy. Third, in the relationship between TMS and team efficacy, only two dimensions of TMS(specialization and task coordination) have an positive Fourth, the one aim of the present study was to investigate the relationships between the three dimensions of TMS(credibility, specialization, and task coordination) and team performance. The results showed the positive relationship between the two dimensions of TMS(credibility and task coordination) and team performance. The specialization factor of TMS is not significantly related with team performance, which is not the unimportance of this factor on team performance, but the importance of credible knowledge, and task coordination referred to the application of this knowledge on problem-solving in order to improve team performance. Fifth, there is a positive impact of team efficacy on team performance. This result indicate the need of team efficacy's enhancement to improve performances under the team system. Finally, the present study found that team efficacy mediates two dimensions of TMS and team performance. This result is the same as the relationship between credibility of TMS and team performance is not significant. Therefore, In the course of TMS it contribute to team performance, Since high team efficacy helps to solve the practical prolems of team such as that two factors of TMS(specialization and task coordination) have a positive effects on team performance, Team effort is need to enhance team efficacy. This study has several important implications for business managers. Firstly, it is important to develop SMM and TMS because it is facilitates the development of team efficacy. The findings suggest that SMM, TMS, and team efficacy alone are not enough to achieve better team performance. Also, Managers should endeavor to creates a climate of interpersonal trust to leverage a team's SMM and TMS. Finally, given the importance of team efficacy on improving a team's performance, team leaders should train team members to acquire teamwork skills that facilitate team efficacy development.

본 연구는 복잡하고 다양한 컨설팅 프로젝트에 참여하는 구성원들의 서로 다른 인식체계와 서로 다른 지식체계에 대 한 프로젝트의 팀성과에 영향을 미치는 팀협업의 연구다. 본 연구의 목적은 컨설팅 프로젝트의 인지적 상호작용 과정에서 팀 공유인지, 팀정보교류, 팀지식통합, 팀효능감의 매개효과를 분석하는 것이다. 본 연구는 선행 연구를 바탕으로 연구모형과 연 구가설을 수립하였다. 컨설팅 프로젝트에 참여한 컨설턴트를 대상으로 자료를 수집하였다. 연구가설에 대한 실증적 분석을 위해 인구통계학적 분석, 타당도 및 신뢰성 분석, 가설검정을 위한 구조모형 분석, 팬텀 변수를 이용한 매개효과 분석을 수행 하였다. 연구 결과, 팀성과를 높이기 위해서는 인식체계인 팀공유인지 및 팀정보교류의 개선이 필요하며, 지식체계인 팀지식 통합이 함께 향상되어야 한다. 따라서 서로 다른 인지 및 지식체계를 통합하고, 서로의 전문성을 신뢰하며, 성공적으로 팀 작 업을 수행할 수 있는 팀효능감이 필요하다. 그리고 향후 인지 과정의 하위요인에 관한 연구가 필요하다고 본다.

This study is a study on the different cognitive systems and different knowledge systems of members participating in complex and diverse consulting projects, and it is a study on team collaboration that affects the team performance of the project. The purpose of this study is to analyze the mediating effects of team shared cognition, team transactive memory, team knowledge integration, and team efficacy in the cognitive interaction process of a consulting project. This study established a research model and research hypothesis based on previous studies. Data were collected from consultants who actually participated in the consulting project. To empirically analyze the research hypothesis, demographic analysis, validity and reliability analysis, structural model analysis for hypothesis verification, and mediating effect analysis using phantom variables were performed. As a result of the study, in order to increase team performance, it is necessary to improve team shared cognition and team transactive memory, which are cognitive systems, and team knowledge integration, which is a knowledge system, must also be improved. Therefore, there is a need for a sense of team efficacy that integrates disparate cognitive and knowledge systems, trusts each other's expertise, and enables successful team work. In addition, future studies on sub-factors of cognitive processes are needed.

Distributed computation follows the models of discrete structures in combinatorial forms. In higher-dimensions, the simplex structures of topological spaces as well as homology are employed to model and analyze distributed asynchronous computations. However, the monotone spaces are the general forms of topological spaces and can be effectively employed to analyze distributed computation. This paper proposes an analytical model of distributed computation in monotone spaces. It is illustrated that, the modeling of distributed computation in monotone spaces helps in determining consistent cuts under closure and convergence of computation. Furthermore, a connective mapping between the simplexes and monotone is constructed.

Programming models for cloud computing has been a research focus recently. Current programming models for cloud computing mainly focus on improving the efficiency of the cloud computing platforms but little has been done on the performance predictability of models. This paper presents a cloud computing programming model, called BSPCloud. The BSPCloud has the advantages of predictable performance. BSPCloud uses a hybrid of distributed-memory and shared-memory bulk synchronous parallel (BSP) programming model. Computing tasks are first divided into a set of coarse granularity bulks which are computed by the distributed-memory BSP model, and each coarse granularity bulk is further divided into a set of bulk threads which are computed by the shared-memory BSP model. BSPCloud makes full use of the multi-core architecture, and more importantly, the performance of the BSPCloud is predictable. We have implemented a proof-of-concept BSPCloud parallel programming library in java. We give the implementation of BSPCloud parallel programming library and show how the library implements hybrid programming. We detail applying the BSPCloud library on matrix multiplication. The performance predictability and speedup are evaluated in the cloud platform. We give the results of experiments.

It is a challenging task to provide an easy–to-use programming interface for distributed shared memory in parallel computing that demands high availability as well as effective synchronization. The current implementations typically use conventional Ethernet as communication medium using broadcast or multicast for synchronization mechanisms. In general, multicast is more efficient than broadcast, as it uses multicast tree to transfer data instead of multicast flooding. Multicast avoids redundant transmissions and lessens network contention. These properties make it significant for distributed shared memory. However, such solution has limited control on response time and process scheduling. This causes uncertainty of time spent in synchronization and interruption of unrelated processes in the same multicast group. In this research, we propose a dynamic SDN multicast mechanism to improve efficiency of synchronization for distributed shared memory by using an SDN-based multicast mechanism and a dynamic core based tree algorithm. The forwarding rule reduces the number of consistency-related messages without interfering the unrelated processes in the same multicast group. The dynamic core-based tree algorithm builds a multicast tree for every multicast group, hereby, SDN controller dynamically adjusts multicast tree according to joining or leaving of nodes. The experiments on the prototype system demonstrate that it could provide better performance compared to the conventional Ethernet based solution in terms of response time and process scheduling in the homogeneous testing environment.

In this paper, two Cortex-A9 processors and microblaze processor are used to build a multiprocessor embedded system. The whole system is connected AXI Interconnect, and the shared memory is used as a communication mechanism. Embedded multiprocessor system is mapped into FPGA, and the feasibility and practicability of the system are verified. Simulation result Indicate that shared memory provides very high performance and low latency access from both processors compared with the DDR memory.

본 논문은 대부분의 Linux 운영체제에서 지원해 주는 System V의 IPC 중 하나인 Shared Memo의 성능을 개선하는 방안을 연구한다. Linux에서 사용되는 Shared Memory는 동일한 메모리 영역에 여러 개의 프로세스가 접근할 수 있도록 해 주는 기술이다. 본 논문에서는 Shared Memory의 큰 두 갈래 중 커널 단계에서 처리 되는 SVR 형식의 Shared Memory를 다룬다. 본 논문에서는 리눅스 운영체제의 공유 메모리 성능 개선 방안을 제안한다. 본 논문에서 제안하는 공유 메모리 성능 개선 방안은 듀얼 코어를 활용하여 기존의 단일 처리기 시스템에서보다 성능을 향상시킬 수 있도록 한다. 공유 메모리를 이용한 프로세스의 동작이 별개의 CPU에서 동작되도록 함으로써 성능 향상을 꾀한다.

본 논문에서는 KDSM(KAIST Distributed Shared Memory) 시스템의 설계 및 구현에 관해서 설명한다. KDSM은 Linux 2.2.13 상에서 실행되는 사용자 수준 라이브러리로 구현되었고, TCP/IP를 기반 통신 구조로 사용한다. KDSM은 페이지 기반 무효화 프로토콜(page-based invalidation protocol)과 다중 쓰기 프로토콜(multiple writer protocol)을 기반으로 하고, HLRC(Home-barred Lazy Release Consistency) 메모리 일관성 모델을 사용한다. KDSM의 성능을 측정하기 위해서 4개의 과학계산용 응용 프로그램을 실행하여 JIAJIA와 성능 비교를 하였다. 그 결과, 2개의 응용은 같은 결과를 냈고, 나머지 2개는 KDSM의 성능이 우수하였다.

In this paper, we give a detailed description of KDSM(KAIST Distributed Shared Memory) system. KDSM is implemented as a user-level library running on Linux 2.2.13, and TCP/IP is used for communication. KDSM uses page-based invalidation protocol, multiple-writer protocol, and supports HLRC(Home-based Lazy Release Consistency) memory consistency model. To evaluate performance of KDSM, we executed 4 scientific applications and compared the result to JLAJLA. The results showed that performance of KDSM almost equal to JIAJIA for 2 applications and performance of KDSM is better than JIAJIA for 2 applications.

The role of the interconnect network, which connects computing nodes to each other, is important in high-performance computing (HPC) systems. In recent years, the peripheral component interconnect express (PCIe) has become a promising interface as an interconnection network for high-performance and cost-effective HPC systems having the features of non-transparent bridge (NTB) technologies. OpenSHMEM is a programming model for distributed shared memory that supports a partitioned global address space (PGAS). Currently, little work has been done to develop the OpenSHMEM library for PCIe-interconnected HPC systems. This paper introduces a prototype implementation of the OpenSHMEM library through a switchless interconnect network using PCIe NTB to provide a PGAS programming model. In particular, multi-interrupt, multi-thread-based data transfer over the OpenSHMEM shared memory model is applied at the implementation level to reduce the latency and increase the throughput of the switchless ring network system. The implemented OpenSHMEM programming model over the PCIe NTB switchless interconnection network provides a feasible, cost-effective HPC system with a PGAS programming model.

통신시스템 소프트웨어는 대단히 크고 복잡하기 때문에, 소프트웨어 재사용성, 하드에어 투명성, 응용에 따른 소프트웨어 재구축의 용이성과 고성능을 위한 컴포넌트 기반 구조를 요구하고 있다. 이런 요구사항을 만족시키기 위하여, 기존 CORBA IIOP의 성능과 통신방식에 대한 분석을 통해 통신시스템에 적합한 공유 메모리 기반의 CORBA 연동 프로토콜을 설계하였다. 설계된 프로토콜은 동일한 인터페이스를 지원하며 동일 시스템 환경에서의 메시지 전송 오버헤드를 최소화시킨다. 다른 프로토콜과의 비교시험 결과 새로운 프로토콜은 약15%-200%의 성능향상을 보여주고 있다. 따라서 본 논문에 제시된 프로토콜은 통신시스템을 위한 CORBA 기반의 컴포넌트 소프트웨어 개발을 위해 사용될 수 있을 것이다.

Since communication systems software is very large and complex, it requires component based architecture for software reusability, hardware transparency, high performance, and easy software reconstruction in different applications. In order to meet these requirements, we analyze performance and inter-process communication techniques of existing CORBA IIOP, and designed a shared memory-based CORBA inter-ORB protocol that would best fit for communication systems software. The designed protocol supports the same interface and can minimize the message transfer overhead in the same host environment. The test results of our protocol compared with other protocols show that the performance is increased by about 15%-200%. We are thus assumed that our protocol can be used in developing CORBA-based component software architecture for communication systems.

Linux에서 사용되는 Shared Memory는 동일한 메모리 영역에 여러 개의 프로세스가 접근할 수 있도록 해 주는 기술이다. 본 논문은 Linux 운영체제에서 지원해 주는 System V의 IPC 중 하나인 Shared Memory를 Dual Core 시스템 상에서 동작하도록 구현한다. 본 논문에서는 커널 단계에서 처리되는 SVR(System V Release) 형식의 Shared Memory를 다룬다. 기존의 공유메모리 방식은 단일 처리기를 이용한 방식이다. 본 논문에서는 dual core를 이용하여 공유메모리 처리를 할 수 있는 시스템을 제안한다. 본 논문에서 제안하는 Dual Core 시스템에서 공유 메모리 기능 구현은 기존의 단일 처리기 시스템에서보다 성능을 향상시킬 수 있도록 한다. 공유 메모리를 이용한 프로세스의 동작이 별개의 CPU에서 동작되도록 함으로써 성능 향상을 꾀한다.

This paper designs Shared Memory on the Dual Core system so that it operates a general System V IPC on the Linux O.S. Shared Memory is the technique that many processes can access to identical memory area. We treat Shared Memory which is SVR in a kernel step. We design a share memory facility of Linux operating system on the Dual Core System. In this paper the suggesting of share memory facility design plan in Dual Core system is enhance the performance in existing an unity processor system as a dual core practical use. We attemp a performance enhance in each CPU for each process which uses a share memory.

본 논문은 대부분의 Linux 운영체제에서 지원해 주는 System V의 IPC 중 하나인 Shared Memory를 Dual Core 시스템 상에서 동작하도록 설계한다. Linux에서 사용되는 Shared Memory는 동일한 메모리 영역에 여러 개의 프로세스가 접근할 수 있도록 해 주는 기술이 다. 본 논문에서는 Shared Memory의 큰 두 갈래 중 커널 단계에서 처리되는 SVR(System V Release) 형식의 Shared Memory를 다룬다. 본 논문에서는 리눅스 운영체제의 공유 메모리 기능을 Dual Core 시스템에서 동작하도록 설계한다. 본 논문에서 제안하는 Dual Core 시스템에서 공유 메모리 기능 설계 방안은 듀얼 코어를 활용하여 기존의 단일 처리기 시스템에서보다 성능을 향상시킬 수 있도록 한다. 공유 메모리를 이용한 프로세스의 동작이 별개의 CPU에서 동작되도록 함으로써 성능 향상을 꾀한다.

This paper designs Shared Memory on the Dual Core system so that it operates a general System V IPC on the Linux O.S. Shared Memory is the technique that many processes can access to identical memory area. We treat Shared Memory in this paper among big two branches of Shared Memory which are SVR in a kernel step format. We design a share memory facility of Linux operating system on the Dual Core System. In this paper the suggesting design plan of share memory facility in Dual Core system is enhancing the performance in existing unity processor system as a dual core practical use. We attempt a performance enhance in each CPU for each process which uses a share memory.

최근에는 고성능 네트웍으로 구성된 클러스터 상에서 사용자 수준 통신을 사용하는 소프트웨어 분산 공유메모리 시스템의 연구가 활발히 진행되고 있다. 본 논문에서는 사용자수준 프로토콜의 표준인 Virtual Interface Architecture(VIA)를 사용하고 확장성 있는 Home-based Lazy release Consistency(HLRC) 모델을 기반으로 하는 소프트웨어 분산공유메모리 시스템을 구현한다. 본 시스템은 VIA의 원격 메모리 쓰기 기능을 최대한 활용하며, 통신 과정에서 통신 버퍼와 사용자 메모리 사이의 복사가 일어나지 않도록 설계되어 높은 성능을 보인다.

본 연구에서는 공유메모리 병렬 컴퓨터 환경에서 N개의 key를 $O(log^2N)$ 시간에 정렬 할 수 있는 병 알고리즘인 SARED-MEMORY-BS와 REDUCED-BS를 설계하였다. REDICED-BS 알고리즘은 각각 로세서에 있는 local memory를 효율적으로 사용할 수 있도록 제안한 parity전략을 사용하였다. 각각의 프로세서에 있는 local memo교를 효율적으로 사용함으로써 REDUCED-BS 알고리즘은 SHARED-MEMORY-BS 알고리즘에 비햐여 통신의 빈도수가 약 1/2정도 감소된 것으로 나타났다. 결과적으로 REDUCED-BS 알고리즘은 병렬 정렬시 통신을 감소시킴으로써 컴퓨터의 사용 효율을 향상시킬 수 있다.

This paper presents parallel sorting algorithm, SHARED-MEMORY-BS and REDUCED-BS, which are implemented on shared-memory parallel computers. These algorithm sort N keys in $O(log^2N)$ time. REDUCED-BS users a parity strategy which gives an idea for the efficient usage of the local memory associated with each processor. By taking advantage of the local memory associated with each processor, the communication of REDUCED-BS is decreased by approximately half that of SHARED-MEMORY-BS. On the basis of alleviating the communication, the algorithm REDUCED-BS results in a significant improvement of performance.

In order to design mutual communication between a distributed control system and a function test facility, we used the Inter-Process Communication(IPC) in two systems and Transmission Control Protocol/Internet Protocol(TCP/IP) protocol. The data from the function test facility are put in the shared memory using an IPC, which is then accessed by the distributed control system through an Application Program Interface(API). The server in the function test facility includes two processes(one for sending and one for receiving), which are generated by the fork function from the client signal. The client in the distributed control system includes two separate programs(one for receiving and one for sending).

Although the concept of "common sense" is often taken for granted, judging whether behavior or knowledge is common sense requires a complex series of mental processes. Additionally, different perceptions of common sense can lead to social conflicts. Thus, it is important to understand how we perceive common sense and make relevant judgments. The present study investigated the dynamics of neural representations underlying judgments of what common sense is. During functional magnetic resonance imaging, participants indicated the extent to which they thought that a given sentence corresponded to common sense under the given perspective. We incorporated two different decision contexts involving different cultural perspectives to account for social variability of the judgments, an important feature of common sense judgments apart from logical true/false judgments. Our findings demonstrated that common sense versus non-common sense perceptions involve the amygdala and a brain network for episodic memory recollection, including the hippocampus, angular gyrus, posterior cingulate cortex, and ventromedial prefrontal cortex, suggesting integrated affective, mnemonic, and social functioning in common sense processing. Furthermore, functional connectivity multivariate pattern analysis revealed that interactivity among the amygdala, angular gyrus, and parahippocampal cortex reflected representational features of common sense perception and not those of non-common sense perception. Our study demonstrated that the social memory network is exclusively involved in processing common sense and not non-common sense. These results suggest that intergroup exclusion and misunderstanding can be reduced by experiencing and encoding long-term social memories about behavioral norms and knowledge that act as common sense of the outgroup.