• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.14 no.3
• /
• pp.339-348
• /
• 2001

More often a commercial package for the structural analysis is necessary in the structural optimum design. In this case the task of combining the package with an optimization program must be required, hut it is not so simple because interchanging some data between them is difficult. Sequential approximate optimization is currently used as a natural way to overcome the hard task. If sequential approximate optimization has wide side constraints that the lower limit of design variables is very small and their upper limit is very large, it is not so easy to obtain approximated functions accurately for the whole design domain. This paper proposes a sequential design domain method, which is very useful to carry out sequential approximate optimization in this case. In this paper, the response surface methodology is used to obtain approximated functions and the orthogonal array is used for design of experiments. The sequential approximate optimization of 3-bar and 10-bar trusses is demonstrated to verify the reliability of the sequential design domain method.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.2
• /
• pp.59-66
• /
• 2013

In this paper, we proposed an optimal ILP algorithm on memory address code generation for DSP in embedded system. This paper using 0-1 ILP formulations DSP address generation units should minimize the memory variable data layout. We identify the possibility of the memory assignment of variable based on the constraints condition, and register the address code which a variable instructs in the program pointer. If the process sequence of the program is declared to the program pointer, then we apply the auto-in/decrement mode about the address code of the relevant variable. And we minimize the loads on the address registers to optimize the data layout of the variable. In this paper, in order to prove the effectiveness of the proposed algorithm, FICO Xpress-MP Modeling Tools were applied to the benchmark. The result that we apply a benchmark, an optimal memory layout of the proposed algorithm then the general declarative order memory on the address/modify register to reduce the number of loads, and reduced access to the address code. Therefor, we proved to reduce the execution time of programs.

• Journal of the Korea Computer Industry Society
• /
• v.5 no.9
• /
• pp.1079-1088
• /
• 2004

Generally, the code motion algorithm accomplishes the run-time optimal connected with the computational optimifation and the register overhead This paper proposes a sparse code motion, which considers the code size, in addition to computational optimization and lifetime optimization. The BCM algorithm carries out the optimal code motion computationally and the LCM algorithm reduces the register overhead in a sparse code motion algorithm. A sparse code motion algorithm is optimum algorithm computationally and lifetime because of suppression unnecessary code motion This algorithm improves runtime and efficiency of the program than the previous work through the performance test.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.3 no.4
• /
• pp.723-736
• /
• 1999

Code optimization phase in a compiler are very important because the phase reduces the running time and the storage size of machine codes. I developed flow analyzers and optimizers on intermediate codes. The flow analyzers generate control-flow and data-flow information. The optimizers transform the intermediate codes into the improved codes using this information. This paper describes the development of flow analyzers and optimizers. I also examined the execution performance, the cost and the dependency of each optimization.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.27 no.3
• /
• pp.40-46
• /
• 1990

This paper describes a design and an implementation of a machine-independent global optimizer which is a required module of RISC compiler system designs. It receives a triple as input and performs data flow analysis, common subexpression elimination and code motion and finally generates the optimized code. Since the implemented optimizer operates on the machine-independent intermediate code, its portability is good for many high level languages and target machines. It performs the effective optimizations to improve the execution time of programs.

• The Korean Journal of Food And Nutrition
• /
• v.25 no.1
• /
• pp.17-25
• /
• 2012

이 연구는 블루베리 즙을 첨가하여 젤리의 제조조건을 최적화하고자 하였다. 16개의 블루베리 즙을 이용한 젤 시료는 Design Expert 프로그램을 이용하여 제조하였으며, 최적화를 위해 블루베리 즙(100~200 g), 설탕(40~160 g), 젤라틴(8~20 g)의 양을 독립변수로, 텍스처, pH, 관능평가 항목을 종속변수로 각각 선정하였다. 반응표면 분석법을 사용하기 위한 실험설계로 중심합성계획을 이용하였다. 각 항목별 최적조건은 Canonical 모형의 수치 최적화(numerical optimization)과 모형적 최적화(graphical optimization)를 통하여 선정하였으며, 그 중 가장 높은 desirability를 갖는 최적점을 선택하여 지점 예측(point prediction)을 통해 도출한 결과, 각 독립변수의 예측된 블루베리 즙을 첨가한 젤리의 최적값은 블루베리 주스 133.63 g, 설탕 160.0 g, 젤라틴은 12.78 g이었다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.12 no.4
• /
• pp.551-561
• /
• 1999

본 논문에서는 쉘 최적화에 대한 연구 결과를 기술하였다. 본 연구의 주목적은 쉘 구조물의 최적형상과 두께 분포를 찾는데 있다. 쉘의 변형에너지를 목적함수로 사용하고 초기 쉘의 부피를 제약조건을 고려하였다. 본 연구에서는 Computer-Aided Geometric Design (CAGD) 기법을 이용하여 쉘의 형상과 그 두께 분포를 표현하였고 쉘의 변형에너지를 측정하기 위해서 가변형 도를 채용한 퇴화 쉘 요소(Degenerated Shell Element)를 도입하였다. 최적 값을 구하기 위해서 세 가지 수학적 프로그래밍 기법을 제공하는 프로그램 DOT를 사용하였다. 마지막으로 새로이 개발된 쉘 최적화시스템의 효율성을 최적화예제로써 증명하였다.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.9-12
• /
• 2024

최근 고성능 컴퓨팅(HPC) 환경에서는 이기종 컴퓨팅 자원의 보편화로 인해 다양한 하드웨어 아키텍처에서의 프로그램 최적화가 필수적이다. 이에 따라 수작업 없이도 하드웨어에 맞춤형 최적화를 자동으로 수행할 수 있는 오토튜닝 기술의 중요성이 커지고 있다. 본 연구에서는 Apache TVM을 활용하여 HPC 워크로드에서의 성능 최적화 가능성을 탐구하였다. 밀집 행렬 연산(GEMM)과 희소 행렬 연산(SpMM)을 주요 워크로드로 설정하고, 국가슈퍼컴퓨터센터의 5호기 누리온에서 실험을 진행하였다. GEMM 연산에서는 누리온 KNL과 Skylake 노드에서 각각 209배, 87배의 성능 향상을 확인하였고, SpMM 연산에서는 BCSR 포맷 기준으로 최대 117배, 10배의 성능 향상이 이루어졌다. 본 연구는 Apache TVM이 인공지능 분야를 넘어 HPC 워크로드에서도 성능 최적화에 유효한 도구임을 시사하며, 향후 다양한 이기종 컴퓨팅 환경에의 적용 가능성을 제시한다.

• Journal of Korea Multimedia Society
• /
• v.10 no.5
• /
• pp.687-697
• /
• 2007

Although the Java bytecode has numerous advantages, there are also shortcomings such as slow execution speed and difficulty in analysis. Therefore, in order for the Java class file to be effectively executed under the execution environment such as the network, it is necessary to convert it into optimized code. We implemented CTOC for transforming to optimized code. In Optimization, SSA Form that distinguish variable by assignment is used to determine value and type statically. Copy propagation, dead code elimination optimization is applied to SSA Form. However, existing SSA Form is related to variable than expression. Therefore, in this paper, to performing optimization to SSA Form expression, after performing copy propagation and dead code elimination, in addition to that, partial redundant expression elimination is performed

• Journal of Energy Engineering
• /
• v.18 no.4
• /
• pp.221-229
• /
• 2009

This study develops the optimized operation program which enables effective and economic operation between individual and connected branch offices by analyzing the current status and influential indicators of district heating companies' capital branch offices. Accordingly, the study examines the efficiency of optimized operation program. In doing so, this study has diagnosed and analyzed various factors, such as boilers, pumps, and relevant tags (temperature, pressure, fuel amount) through investigation of individual branch offices, and finally succeeded in developing wide-ranging data base by factor covering one-year time period. Additionally, after running the optimized operation program, different branch offices, optimum preference has turned out "incinerator receiving heat from KEPCO>CHP >PLBs>PLBw." Meantime, except the connected offices, there has been no big difference between actual and optimum operation program in branch offices. Meanwhile, the integrated optimum operation program has made it possible the most optimal result only via the connecting supply and demand heat without changing received Heat from KEPCO which is the same as total productive heat. The result has showed that the reduction percentage per day is 2.45~6.80%, and the reduction cost per day is 22,727~60,077 thousand won given the randomly selected sample days. In particular, winter time shows the highest demand with the largest reduction cost whereas summer time illustrates the lowest demand with the smallest reduction cost. Given this result, reduction cost per year compared to actual heat production cost for one year theoretically would be 84 hundred million won. Also, the economic effect showed that the reduction cost percentage per year is more than 2.74% on heat production cost per year for all capital branch offices.