• The Journal of the Korea Contents Association
• /
• v.5 no.4
• /
• pp.204-211
• /
• 2005

This paper introduces three loop splitting methods such as minimum dependence distance method, Polychronopoulous' method, and first dependence method for exploiting parallelism from single loop which already developed. And it also Indicates their several problems. We extend the first dependence method which is the most effective one among three loop splitting methods, and propose more powerful loop splitting method to enhance parallelism on single loop. The proposed algorithm solves several problems, such as anti-flow dependence and g=gcd(a,c) > 1, that the first dependence method has.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.565-569
• /
• 2006

This paper reviews some loop partitioning techniques such as loop splitting method by thresholds and Polychronopoulos' loop splitting method for exploiting parallelism from single loop which already developed. We propose improved loop splitting method for maximizing parallelism of single loops with non-constant dependence distances. By using the distance for the source of the first dependence, and by our defined theorems, we present generalized and optimal algorithms for single loops with non-uniform dependences. The algorithms generalize how to transform general single loops into parallel loops.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.1
• /
• pp.23-33
• /
• 2000

We propose a model which overlaps communication with computation for efficient communication in the data-parallel programming paradigm. The overlapping model divides a given loop partition into several sub-partitions to obtain computation which can be overlapped with communication. A loop partition sometimes refers to other data partitions, but not all iterations in the loop partition require non-local data. So, a loop partition may be divided into a set of loop iterations which require non-local data, and a set of loop iterations which do not. Each loop sub-partition is dynamically scheduled depending on associated message arrival, The experimental results for a few benchmarks in IBM SP2 show enhanced performance in our overlapping model.

• Journal of the Korea Convergence Society
• /
• v.6 no.3
• /
• pp.51-57
• /
• 2015

This paper proposes an efficient method such as Extended Three Region Partitioning Method for nested loops with irregular dependences for maximizing parallelism. Our approach is based on the Convex Hull theory, and also based on minimum dependence distance tiling, the unique set oriented partitioning, and three region partitioning methods. In the proposed method, we eliminate anti dependences from the nested loop by variable renaming. After variable renaming, we present algorithm to select one or more appropriate lines among given four lines such as LMLH, RMLH, LMLT and RMLT. If only one line is selected, the method divides the iteration space into two parallel regions by the selected line. Otherwise, we present another algorithm to find a serial region. The selected lines divide the iteration space into two parallel regions as large as possible and one or less serial region as small as possible. Our proposed method gives much better speedup and extracts more parallelism than other existing three region partitioning methods.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.06a
• /
• pp.16-19
• /
• 2001

일반적인 응용 프로그램에서 루프는 대부분의 수행시간을 차지하기 때문에 병렬성 추출의 핵심 부분이라 할 수 있다. 병렬성이 많은 구조는 루프 구조이며, 루프를 병렬로 처리하기 위해 각 반복간에 존재하는 데이터의 종속은 프로세서간의 동기화가 필요하다. 본 논문에서는 다중첨자를 갖는 1차원 배열의 루프의 병렬화를 위해 다수 개의 동일한 종속값을 이용하여 종속함수를 생성하고 이를 이용하여 종속관계가 성립하지 않는 비종속 구간(Non-dependence part)을 구한다. 그리고 동일한 값을 가지는 복수개의 종속값 간의 동기화는 외부루프 분할 기법을 이용하여 적은 횟수의 청자가 외부에 위치하도록 하여 간소화한 후 단일 첨자를 갖는 루프에 동기화를 수행하는 기법을 제시한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.04a
• /
• pp.239-242
• /
• 2001

일반적인 응용 프로그램에서 병렬성이 많은 구조는 루프 구조이며, 루프를 병렬로 처리하기 위해 동기화가 필요하다. 본 논문에서는 다중첨자를 갖는 1차원 배열의 루프의 병렬화를 위해 다수 개의 동일한 종속값을 이용하여 종속함수를 생성하고 이를 이용하여 종속관계가 성립하지 않는 비종속 구간(Non-dependence part)을 구한다. 그리고 동일한 값을 가지는 복수개의 종속값 간의 동기화는 외부루프 분할 기법을 이용하여 간소화 한 후 단일 첨자를 갖는 루프에 동기화를 수행하는 기법을 제시한다.

• Journal of Korean Society of Transportation
• /
• v.22 no.6
• /
• pp.109-120
• /
• 2004

Since the late of 1990, there have been number of studies on the required number of probe vehicles and/or optimal aggregation interval sizes for travel time estimation and forecasting. However, in general one to five minutes are used as aggregation intervals for the travel time estimation intervals for the travel time estimation and/or forecasting of loop detector system without a reasonable validation. The objective of this study is to deveop models for identifying optimal aggregation interval sizes of loop detector data for travel time estimation and prediction. This study developed Cross Valiated Mean Square Error (CVMSE) model for the link and route travel time forecasting, The developed models were applied to the loop detector data of Kyeongbu expressway. It was found that the optimal aggregation sizes for the travel time estimation and forecasting are three to five minutes and ten to twenty minutes, respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.9B
• /
• pp.751-759
• /
• 2003

We analyze the performances, such as the cell loss probability(CLR) and the cell delay time, of time division multiplexed(TDM) optical output buffers using traveling delay lines or delay-line loops for buffering. Since traveling delay lines used for buffering are superior over delay-line loops in terms of simplicity and signal quality, they were used in a conventional TDM optical output buffer. However, the latter is more flexible than the former in that the cell storage time is adjustable by changing the recirculating times of a cell in the loops. So we propose a novel TDM optical output buffer using delay-line loops for buffering. We show that the proposed TDM optical output buffer can reduce the number of buffering unit required to achieve a CLR of less than 10$^{-9}$ . When the number of buffering unit is sufficiently large, we show that both TDM optical output buffers have same cell delay time characteristic.

• Proceedings of the KIPE Conference
• /
• 2019.11a
• /
• pp.152-153
• /
• 2019

본 논문에서는 캐패시터-분할 타입 능동전력 디커플링 회로를 위한 전압제어 방법을 제안한다. 능동전력 디커플링 회로는 시스템에 필요한 커패시턴스를 낮추어 전해커패시터를 필름 커패시터로 대체하여 시스템 수명과 전력밀도를 높일 수 있는 장점이 있다. 그러나 일반적으로 오픈 루프 제어방식의 전압제어 방식을 사용하여 파라미터 값의 변화에 민감하다는 단점을 가지고 있다. 이에 본 논문에서는 커패시터-분할 타입 능동전력 디커플링 회로를 위한 폐루프 제어 방법을 제안한다.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1987.10a
• /
• pp.165-172
• /
• 1987