Search results for: Sung-Ho Han

Authors: Kwangho Cha, Hyeyoung Cho, Sungho Kim

Abstract:

As many scientific applications require large data processing, the importance of parallel I/O has been increasingly recognized. Collective I/O is one of the considerable features of parallel I/O and enables application programmers to easily handle their large data volume. In this paper we measured and analyzed the performance of original collective I/O and the subgroup method, the way of using collective I/O of MPI effectively. From the experimental results, we found that the subgroup method showed good performance with small data size.

Keywords: Collective I/O, MPI, parallel file system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1575

Authors: Hyeyoung Cho, Sungho Kim, Sik Lee

Abstract:

I/O workload is a critical and important factor to analyze I/O pattern and to maximize file system performance. However to measure I/O workload on running distributed parallel file system is non-trivial due to collection overhead and large volume of data. In this paper, we measured and analyzed file system activities on two large-scale cluster systems which had TFlops level high performance computation resources. By comparing file system activities of 2009 with those of 2006, we analyzed the change of I/O workloads by the development of system performance and high-speed network technology.

Keywords: I/O workload, Lustre, GPFS, Cluster File System

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1461

Authors: Jihyung Kim, Taejin Kim, Sungho Park, Jun-Dong Cho

Abstract:

In this paper, the problem of reducing switching activity in on-chip buses at the stage of high-level synthesis is considered, and a high-level low power bus binding based on dynamic bit reordering is proposed. Whereas conventional methods use a fixed bit ordering between variables within a bus, the proposed method switches a bit ordering dynamically to obtain a switching activity reduction. As a result, the proposed method finds a binding solution with a smaller value of total switching activity (TSA). Experimental result shows that the proposed method obtains a binding solution having 12.0-34.9% smaller TSA compared with the conventional methods.

Keywords: bit reordering, bus binding, low power, switching activity matrix

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1304

Authors: Jihyung Kim, Taejin Kim, Sungho Park, Jun-Dong Cho

Abstract:

In this paper, we address the problem of reducing the switching activity (SA) in on-chip buses through the use of a bus binding technique in high-level synthesis. While many binding techniques to reduce the SA exist, we present yet another technique for further reducing the switching activity. Our proposed method combines bus binding and data sequence reordering to explore a wider solution space. The problem is formulated as a multiple traveling salesman problem and solved using simulated annealing technique. The experimental results revealed that a binding solution obtained with the proposed method reduces 5.6-27.2% (18.0% on average) and 2.6-12.7% (6.8% on average) of the switching activity when compared with conventional binding-only and hybrid binding-encoding methods, respectively.

Keywords: low power, bus binding, switching activity, multiple traveling salesman problem, data sequence reordering

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1333

Authors: Jihyung Kim, Taejin Kim, Sungho Park, Jun-Dong Cho

Abstract:

A conventional binding method for low power in a high-level synthesis mainly focuses on finding an optimal binding for an assumed input data, and obtains only one binding table. In this paper, we show that a binding method which uses multiple binding tables gets better solution compared with the conventional methods which use a single binding table, and propose a dynamic bus binding scheme for low power using multiple binding tables. The proposed method finds multiple binding tables for the proper partitions of an input data, and switches binding tables dynamically to produce the minimum total switching activity. Experimental result shows that the proposed method obtains a binding solution having 12.6-28.9% smaller total switching activity compared with the conventional methods.

Keywords: low power, bus binding, switching activity, multiplebinding tables

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1184

Authors: Hyeyoung Cho, Sungho Kim, SangDong Lee

Abstract:

I/O workload is a critical and important factor to analyze I/O pattern and file system performance. However tracing I/O operations on the fly distributed parallel file system is non-trivial due to collection overhead and a large volume of data. In this paper, we design and implement a parallel file system logging method for high performance computing using shared memory-based multi-layer scheme. It minimizes the overhead with reduced logging operation response time and provides efficient post-processing scheme through shared memory. Separated logging server can collect sequential logs from multiple clients in a cluster through packet communication. Implementation and evaluation result shows low overhead and high scalability of this architecture for high performance parallel logging analysis.

Keywords: I/O workload, PVFS, I/O Trace.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1560

Authors: Sungho Lee, Goonjae Lee, Chaeyeon Lim, Sunkuk Kim

Abstract:

Foundation of tower crane serves to ensure stability against vertical and horizontal forces. If foundation stress is not sufficient, tower crane may be subject to overturning, shearing or foundation settlement. Therefore, engineering review of stable support is a highly critical part of foundation design. However, there are not many professionals who can conduct engineering review of tower crane foundation and, if any, they have information only on a small number of cranes in which they have hands-on experience. It is also customary to rely on empirical knowledge and tower crane renter-s recommendations rather than designing foundation on the basis of engineering knowledge. Therefore, a foundation design automation system considering not only lifting conditions but also overturning risk, shearing and vertical force may facilitate production of foolproof foundation design for experts and enable even non-experts to utilize professional knowledge that only experts can access now. This study proposes Automatic Design Algorithm for the Tower Crane Foundations considering load and horizontal force.

Keywords: Tower Crane, Automatic Design, Foundations, Optimization Algorithm, Stability

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 7213

Authors: Sungho Kim, Chaehoon Park, Yukyung Choi, Soon Kwon, In So Kweon

Abstract:

In this paper, a novel corner detection method is presented to stably extract geometrically important corners. Intensity-based corner detectors such as the Harris corner can detect corners in noisy environments but has inaccurate corner position and misses the corners of obtuse angles. Edge-based corner detectors such as Curvature Scale Space can detect structural corners but show unstable corner detection due to incomplete edge detection in noisy environments. The proposed image-based direct curvature estimation can overcome limitations in both inaccurate structural corner detection of the Harris corner detector (intensity-based) and the unstable corner detection of Curvature Scale Space caused by incomplete edge detection. Various experimental results validate the robustness of the proposed method.

Keywords: Feature, intensity, contour, hybrid.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1830

Authors: Youkyung Park, SeungYup Kang, SungHo Kim, SimKyun Yook

Abstract:

IMCS is Integrated Monitoring and Control System for thermal power plant. This system consists of mainly two parts; controllers and OIS (Operator Interface System). These two parts are connected by Ethernet-based communication. The controller side of communication is managed by CNet module and OIS side is managed by data server of OIS. CNet module sends the data of controller to data server and receives commend data from data server. To minimizes or balance the load of data server, this module buffers data created by controller at every cycle and send buffered data to data server on request of data server. For multiple data server, this module manages the connection line with each data server and response for each request from multiple data server. CNet module is included in each controller of redundant system. When controller fail-over happens on redundant system, this module can provide data of controller to data sever without loss. This paper presents three main features – separation of get task, usage of ring buffer and monitoring communication status –of CNet module to carry out these functions.

Keywords: Ethernet communication, DCS, power plant, ring buffer, data integrity

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1563