• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.6
• /
• pp.56-62
• /
• 2009

Inexpensive high performance computer systems combined with high speed networks and machine independent communication libraries have made cluster computing a viable option for parallel applications. In a heterogeneous cluster environment, efficient resource management is critically important since the computing power of the individual computer system is a significant performance factor when executing applications in parallel. This paper presents a dynamic task manager, called DWM (dynamic work manager). It makes a heterogeneous cluster system fully utilize the different computing power of its individual computer system. We measure the performance of DWM in a heterogeneous cluster environment with several kernel-level benchmark programs and their programming complexity quantitatively. From the experiments, we found that DWM provides competitive performance with a notable reduction in programming effort.

• Journal of Computing Science and Engineering
• /
• v.6 no.4
• /
• pp.287-293
• /
• 2012

We introduce a methodology for the study of the application-level performance of time-sharing parallel jobs on a set of compute nodes in high performance clusters and report our findings. We assume that parallel jobs arriving at a cluster need to share a set of nodes with the jobs of other users, in that they must compete for processor time in a time-sharing manner and other limited resources such as memory and I/O in a space-sharing manner. Under the assumption, we developed a methodology to simulate job arrivals to a set of compute nodes, and gather and process performance data to calculate the percentage slowdown of parallel jobs. Our goal through this study is to identify a better combination of jobs that minimize performance degradations due to resource sharing and contention. Through our experiments, we found a couple of interesting behaviors for overlapped parallel jobs, which may be used to suggest alternative job allocation schemes aiming to reduce slowdowns that will inevitably result due to resource sharing on a high performance computing cluster. We suggest three job allocation strategies based on our empirical results and propose further studies of the results using a supercomputing facility at the San Diego Supercomputing Center.

• Journal of Digital Contents Society
• /
• v.10 no.2
• /
• pp.249-258
• /
• 2009

As wireless internet markets become larger than before, many mobile applications are also being developed actively. In this circumstances mobile devices such as cell phones, PDAs are playing an important role to satisfy the user's need of ubiquitous computing. Due to the hardware limitations, however, the mobile devices like PDA can not run large-scale softwares by itself. The main goal of this paper is to make large-scale applications runnable on PDA. To accomplish this, we used the PDA-JPVM cluster computing engine which has been already developed by us. We found out by running the applications and the performance evaluation that large-scale Java softwares can easily run on the hardware-limited PDA. And the performance evaluation results are also presented.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.5
• /
• pp.385-393
• /
• 2004

Cluster systems have been widely used for solving problems in various application domains, and regarded as useful high performance computing resources. As the number of cluster system user is increasing, it is no less important to maintain stable operation than to improve cluster system performance. Although hardware preventive maintenance is important for keeping normal operation, the testing tool which can be used for general cluster systems during maintenance has received little attention. In this Paper, considering hardware Preventive maintenance, we suggest a testing tool for hardware of cluster system. The cluster system testing tool which is named CTS(Cluster system Test Suite) has two check routines; one for memory, and the other for NIC respectively. The CTS is designed to support the common features of general cluster systems and all the Jobs such as setting test conditions to querying the results can be done entirely within an integrated GUI environment. CTS is used as the testing tool for two kinds of cluster systems during maintenance, and the experimental results show that CTS reports useful information for cluster systems management.

• Journal of information and communication convergence engineering
• /
• v.18 no.3
• /
• pp.183-187
• /
• 2020

The general Wi-Fi network connection structure is that a number of IoT (Internet of Things) sensor nodes are directly connected to one AP (Access Point) node. In this structure, the range of the network that can be established within the specified specifications such as the range of signal strength (RSSI) to which the AP node can connect and the maximum connection capacity is limited. To overcome these limitations, multiple middleware bridge technologies for dynamic scalability and load balancing were studied. However, these network expansion technologies have difficulties in terms of the rules and conditions of AP nodes installed during the initial network deployment phase In this paper, an intelligent edge computing IoT device is developed for constructing an intelligent autonomous cluster edge computing network and applying it to real-time road danger context aware and notification system through an intelligent risk situation recognition algorithm.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.1098-1100
• /
• 2002

Inter-RNC(radio network controller) and inter-BSC(base station center) hard handovers involve rerouting. Thus additional delay time for the new bearer setup and reroute determination should be considered. For theses kinds of handovers we propose a new scheme and coined it as ICHP(Inter-Cluster Handover Prediction). When a call approaches a boundary cell of a cluster, ICHP prepares for the handover by performing resource reservations and reroute computation in advance. Performance analysis shows that ICHP provides faster handover and that it lowers the blocking probability.

• Journal of computational fluids engineering
• /
• v.6 no.3
• /
• pp.51-56
• /
• 2001

As IT(Information Technology) has been developing, an application engineering is advanced so quickly. Especially, CFD field that is influenced greatly by Computing Power is an outstanding example. In this paper, it says a research tendency of the KISTI Supercomputing Center that performs the CFD research based on IT. The representative researches are the National Grid Project, TeraCluster Construction and development and a supporting plan for Supercomputer users' parallelization.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.31 no.4
• /
• pp.16-25
• /
• 2003

A parallel computing environment based on Windows operating system was constructed and a performance test was mode in comparison with Linux based systems. The Windows 2000 cluster was composed with servers and clients connected by Fast-ethernet, within which two sub-clusters may operates together or separately. Compaq Visual Fortran complier and two MPI libraries, MPICH.NT.1.2.2 and NT-MPICHNT.1.2 were installed as computing tools. Parallel computing performance tests were carried out using two-dimensional preconditioned Navier-Stokes code to examine the dependency on the number of processors, problem size and MPI libraries, those were compared with results from Linux clusters. Results shows that a cluster based on the user-friendly Windows operating system is also useful for the parallel computing and has good performance comparable to the previous Linux clusters.

• The KIPS Transactions:PartA
• /
• v.13A no.2 s.99
• /
• pp.157-162
• /
• 2006

Recent High Performance Computing (HPC) platforms can be classified as Shared-Memory Multiprocessors (SMP), Massively Parallel Processors (MPP), and Clusters of computing nodes. These platforms are deployed in many scientific and engineering applications which require very high demand on computing power. In order to realize an optimal performance for these applications, it is crucial to find and use the suitable computing platforms and programming paradigms. In this paper, we use SPEC HPC 2002 benchmark suite developed in various parallel programming models (MPI, OpenMP, and hybrid of MPI/OpenMP) to find an optimal computing environments and programming paradigms for them through their performance analyses.

• 한국전산유체공학회:학술대회논문집
• /
• 2002.05a
• /
• pp.1-1
• /
• 2002

This abstract contains information of PC-Cluster systems that Samsung Advanced Institute of Technology (SAIT) have developed and researched since 1998. This paper introduce not only the reasons that why SAIT has been interested and focused in such system, but also where to SAIT has applied. There are many other applicable areas but this paper emphasis only in the fields of high performance computing and storage system. With the result of SAIT's PC-Cluster research, we hope we can create many new applicable areas with collaborations.