• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.115-117
• /
• 2002

객체지향 프로그램 슬라이싱을 위한 Loren D Larsen and Marry Jean Harrold가 제안하는 방법은 절차적 프로그램 슬라이싱을 위한 시스템 종속성 그래프 표현방법에 객체지향 패러다임을 표현할 수 있도록 확장하며 2단계 마킹 알고리즘을 적용하여 슬라이스를 계산한다. 시스템 종속성 그래프를 이용한 슬라이싱 방법은 클래스 멤버 변수와 전역변수 및 인스턴스 변수에 대하여 각 메소드 호출 및 진입정점에 actual_in, actual_out, formal_in, formal_out 정점들이 추가되어 복잡도가 증가한다. 본 논문에서는 이들 변수를 클래스 정점의 멤버간선으로 연결하여 각 메소드의 문장에서 사용하면 진출간선 정의하면 진입간선으로 연결하여 정점과 간선들의 개수를 최소화할 수 있도록 시스템 종속성 그래프를 개선하였다. 제안하는 시스템 종속성 그래프는 그래프 복잡도의 최소화와 2단계 알고리즘에 의한 정확한 슬라이스 계산이 장점이다. C++ 예제 프로그램을 적용하여 그래프 복잡도의 감소와 정확한 슬라이스 계산을 기존의 방법과 비교하여 개선됨을 확인하였다.

• Journal of the Korea Computer Industry Society
• /
• v.5 no.5
• /
• pp.711-720
• /
• 2004

There are conversion of expression motion and assignment motion inclusive espression motion in techniques that optimized program computationally and livelly. In this paper, I suggest that an algorithms of reconstruct unnecessary code motion which is improves Knoop's algorithms that have ambiguous. It is occurred by mixing the node level analysis and the instruction level analysis. This algorithm improves runtime and efficiency of a program by avoiding the unnecessary recalculations and reexecutions of expressions and assignment statements.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11a
• /
• pp.321-324
• /
• 2003

본 논문에서는 코드 최적화를 위하여 계산적으로나 수명적으로 제한이 없는 희소 코드 모션 알고리즘을 제안한다. 이 알고리즘은 지나친 레지스터의 사용을 막기 위하여 불필요한 코드 모션을 억제한다. 또한, 본 논문에서는 기존 알고리즘의 술어의 의미가 명확하지 않은 것을 개선하였고 노드 단위 분석과 명령어 단위 분석을 혼용했기 때문에 발생하는 모호함도 개선하였다. 따라서, 제안한 알고리즘은 불필요하게 중복된 수식이나 배정문의 수행을 피하게 함으로써, 프로그램의 불필요한 재계산이나 재실행을 하지 않게 하여 프로그램의 능률 및 실행시간을 향상시킨다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.10a
• /
• pp.754-757
• /
• 2007

This paper presents automatic LCD gamma control system using gamma curve optimization. It controls automatically gamma adjustment registers in mobile LCD driver IC to reduce average gamma error and adjusting time. The proposed gamma system contains module-under-test (MUT, LCD module), PC installed with program, multimedia display tester for measuring luminance, and control board for interface between PC and LCD module. We have developed a new algorithm using 6-point programmable matching technique with reference gamma curve. Developed algorithm and program are generally applicable for most of the LCD modules.

• Journal of Internet Computing and Services
• /
• v.12 no.3
• /
• pp.131-137
• /
• 2011

GPUs were originally designed for graphic processing, and GPGPUs are general-purpose GPUs for numerical computation with high performance and low electric power. In this paper, we implemented the parallel LU factorization program for GPGPUs. In CUDA, which is computational environment for Nvidia GPGPUs, domains are divided into blocks, and multi-threads compute each sub-blocks Simultaneously. In LU factorization program, computation order should be artificially decided due to the data dependence. To resolve the data dependancy, we suggested a parallel LU program for GPGPUs, and also explained parallel reduction algorithm for partial pivoting of LU factorization. We finally present performance analysis to show efficiency of the parallel LU factorization program based on multi-threads on GPGPUs.

• Journal of KIISE:Software and Applications
• /
• v.37 no.12
• /
• pp.882-891
• /
• 2010

Many automated dynamic test data generation technique have been proposed. The techniques evaluate fitness of test data through executing instrumented Software Under Test (SUT) and then generate new test data based on evaluated fitness values and optimization algorithms. Previous researches and experiments have been showed that these techniques generate effective test data. However, optimization algorithms in these techniques incur much time to generate test data, which results in huge test case generation cost. In this paper, we propose a technique for reducing the time of evaluating a fitness of test data among steps of dynamic test data generation methods. We introduce the concept of Fitness Evaluation Program (FEP), derived from a path constraint of SUT. We suggest a test data generation method based on FEP and implement a test generation tool, named ConGA. We also apply ConGA to generate test cases for C programs, and evaluate efficiency of the FEP-based test case generation technique. The experiments show that the proposed technique reduces 20% of test data generation time on average.

• Journal of Internet Computing and Services
• /
• v.9 no.1
• /
• pp.55-67
• /
• 2008

This paper presents the assignment motion algorithm unrestricted for code optimization computationally. So, this algorithm is suppressed the unnecessary code motion in order to avoid the superfluous register pressure, we propose the assignment motion algorithm added to the final optimization phase. This paper improves an ambiguous meaning of the predicate. For mixing the basic block level analysis with the instruction level analysis, an ambiguity occurred in Knoop's algorithm. Also, we eliminate an ambiguity of it. Our proposal algorithm improves the runtime efficiency of a program by avoiding the unnecessary recomputations and reexecutions of expressions and assignment statements.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.2
• /
• pp.19-26
• /
• 2010

The Windows registry is a hierarchical database where the configuration data of a system or application programs is stored. In this paper, we presented and implemented a registry virtualization algorithm and measured its performance. The registry virtualization algorithm presented in the paper is called Copy-One-level On Write-Open(COOWO) that is a modified version of general Copy On Whte(COW) method to make it suitable for registry virtualization. In this paper, we implemented the proposed algorithm as a dynamically loadable library in Windows and applied it to many Windows application programs. This paper is meaningful since we described a registry virtualization algorithm in detail in situation where we can not find papers that describe the registry virtualization in detail, and we could find the performance of the algorithm can be used in the real applications.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04a
• /
• pp.103-105
• /
• 2001

디스크 I/O 성능 개선을 위한 여러 캐슁과 선반입 알고리즘이 연구되어져 왔다. 선반입 알고리즘은 디스크 접근 횟수를 줄임으로써 시스템의 프로세싱 시간을 줄여준다. 이는 선반입 기법을 통해 응용 프로그램이 실제로 데이터 블록을 요청하기 전에 시스템이 데이터 블록을 읽도록 하기 때문이다. 이 논문은 OBA 선반입 기법의 확장판인 E-OBA 선반입 기법을 제안한다. 제안된 방법은 디스크 대역폭의 여유가 허락된다면 연속적으로 데이터 블록을 선반입 하여 시스템 프로세싱 시간을 줄인다. 그리고 기존의 캐슁되어 있는 데이터 블록을 보호하기 위해 선반입 하는 데이터 블록의 개수를 제한한다. 특별히, 제안된 선반입 정책은 순차적으로 접근되는 큰 파일에 대해서 시스템 성능을 개선시킨다.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.18 no.2
• /
• pp.30-37
• /
• 2014

In this study, we studied how to perform the structural analysis which need a large-capacity flash memory with the computer program when the flash memory storage of a personal computer has no enough room for the analysis of structure. As one of the solutions of this problem, the blocking method of stiffness matrix, which is a method that stiffness matrix is divided by a few blocks and each block is sequentially used for the calculation of matrix decomposition, is proposed and an algorithm available in computer program is derived on the method. Finally, A structural analysis program (sNs) based on this study is developed and the correctness and efficiency of the algorithm is founded through some examples which are fundamental in structural analysis.