• Journal of the Korea Society of Computer and Information
• /
• v.18 no.4
• /
• pp.1-11
• /
• 2013

Dynamic Software Update(DSU) is a technique, which updates a new version of the software to a running process without stopping. Many DSU systems that update C application programs are introduced. However, these DSU systems differ in implementation method or in main features. In this paper, we propose a new DSU system that can solve some disadvantages of existing DSU systems. DSU system presented in this paper splits existing program to code, global data and local data and then updates each part of the program considering the characteristics of the respective parts. The proposed system in this paper is implemented and tested on Linux. Also, we compared our DSU system with other DSU systems and we could find some strength of our DSU system. First, the code memory usage of our DSU system can be efficient since our system does not need to maintain code of an old version. Second, the global data memory waste is small because our system does not need to allocate the global data again which is not modified in the new version. Finally, we restore local data of old version in stack area of the new version using stack reconstruction technique. This paper is meaningful since we proposed a new DSU method and we implemented a full DSU system using the method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.2C
• /
• pp.181-186
• /
• 2004

Recently, there has been intensive focus on Turbo Product Codes(TPCs) which have low decoding complexity and achieve near-optimum performances at high code-rate. This paper present a parallel algorithm of turbo product codes enable simultaneous decoding of row and column. The row and column decoders operate in parallel and update each other after row and column has been decoded. simulation results show that the performance of proposed parallel turbo code is almost the same as that conventional scheme for several turbo product codes.

• Computers and Concrete
• /
• v.7 no.2
• /
• pp.145-158
• /
• 2010

In this work the implementation of a production-level port of the Microplane constitutive model for concrete into the EPIC code is described. The port follows guidelines outlined in the Material Model Module (MMM) standard used in EPIC to insure a seamless interface with the existing code. Certain features of the model were not implemented using the MMM interface due to compatibility reasons; for example, a separate module was developed to initialize, store and update internal state variables. Objective strain and deformation measures for use in the material model were also implemented into the code. Example calculations were performed and illustrate the veracity of this new implementation.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.8
• /
• pp.808-812
• /
• 2008

In recent years, the software vulnerability becomes an important problem to the safety in operating system kernel. Many organizations endeavor to publish patches soon after discovery of vulnerability. In spite of the effort, end-system administrators hesitate to apply the patches to their system. The reasons of hesitation are the reboot disruption and the distrust of patches. To solve this problem we propose a dynamic update system for non-updatable kernel, named DUNK. The DUNK provides: 1) a dynamic update mechanism at function-level granularity to overcome the reboot disruption and 2) a safety verification mechanism to overcome the distrust problem, named MAFIA. In this paper, we describe the design of DUNK and detailed algorithm of MAFIA.

• The Bulletin of The Korean Astronomical Society
• /
• v.38 no.2
• /
• pp.101.2-101.2
• /
• 2013

We present a new computational fluid dynamic (CFD) simulation code. The code employs the moving and polyhedral unstructured mesh scheme, which is known as a superior approach to the conventional SPH (smoothed particle hydrodynamics) and AMR (adaptive mesh refinement) schemes. The code first generates unstructured meshes by the Voronoi tessellation at every time step, and then solves the Riemann problem for surfaces of every Voronoi cell to update the hydrodynamic states as well as to move former generated meshes. For the second-order accuracy, the MUSCL-Hancock scheme is implemented. To increase efficiency for generating Voronoi tessellation we also develop the incremental expanding method, by which the CPU time is turned out to be just proportional to the number of particles, i.e., O(N). We will discuss the applications of our code in the context of cosmological simulations as well as numerical experiments for galaxy formation.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.2
• /
• pp.65.2-65.2
• /
• 2014

We present a new hydrodynamic simulation code based on the Voronoi tessellation for estimating the density precisely. The code employs both of Lagrangian and Eulerian description by adopting the movable mesh scheme, which is superior to the conventional SPH (smoothed particle hydrodynamics) and AMR (adaptive mesh refinement) schemes. The code first generates unstructured meshes by the Voronoi tessellation at every time step, and then solves the Riemann problem for all surfaces of each Voronoi cell so as to update the hydrodynamic states as well as to move current meshes. Besides, the IEM (incremental expanding method) is devised to compute the Voronoi tessellation to desired degree of speed, thereby the CPU time is turned out to be just proportional to the number of particles, i.e., O(N). We discuss the applications of our code in the context of cosmological simulations as well as numerical experiments for galaxy formation.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10a
• /
• pp.233-234
• /
• 2007

• Journal of Korea Multimedia Society
• /
• v.14 no.7
• /
• pp.949-963
• /
• 2011

Generally, It is indispensible to use an ISP(In System Programming) tool for upgrading, patching, or changing the system software of the each sensor nodes in wireless sensor networks. While under a small number of nodes, the upgrading task is not a serious burden, however if there are a large number of nodes to be updated, the task is almost impossible to do for the given constrains such as limited budgets and resources. Based on this observation, in this paper we have proposed a novel upgrading scheme based on a single hop in IEEE 802.15.4 PAN(Personal Area Network)s. Simulation results have shown the scheme outcomes the conventional methods in the performance measures.

• Proceedings of the IEEK Conference
• /
• 2001.06a
• /
• pp.205-208
• /
• 2001

According to the IMT-2000 specification of 3GPP(3rd Generation Partnership Project) and 3GPP2, Turbo codes is selected as a FEC(forward error correction) code for even higher reliable data communication. In 3GPP complied IMT-2000 system, channel coding under consideration is the selective use of convolutional coding and Turbo codes of 1/3 code rate with 4 constraint length. Suggesting a new path metric normalization method, we achieved a low complexity and high performance SOVA decoder for Turbo Codes, Further more, we analyze the decoding performance with respect to update depth and find out the optimal value of it by using computer simulation. Based on the simulation result, we designed a SOVA decoder using VHDL and implemented it into the Altera EPF10K100GC503FPGA.

• Nuclear Engineering and Technology
• /
• v.15 no.1
• /
• pp.41-49
• /
• 1983

A computer code, BERD (Bayesian Estimation of Reliability Data), has been developed and tested in order to update the data for the reliability analysis of safety related systems in a specific nuclear power plant. The code has been used to derive the plant-specific data for reliability analysis of the auxiliary feedwater system of a pressurized water reactor. The prior information for components selected was taken from the U.S. Reactor Safety Study, WASH-1400, and the operating experiences from published licensee event reports. The results show that the updated data are well fitted to log-normal distribution curves and the error factors are reduced significantly.