• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10c
• /
• pp.149-153
• /
• 2006

외부 조인은 조인에 참여하는 릴레이션들에서 조인 조건을 만족하지 않는 한쪽 또는 양쪽 튜플들도 결과로 반환하는 조인 연산으로 OLAP 질의 처리, 계층적 뷰 처리, 중첩 질의 처리 등의 다양한 고급 데이타베이스 응용에서 널리 사용된다. 많은 상용 DBMS에서 외부 조인을 지원하고 있으나, 상세한 구현 방법은 공개되어 있지 않다. 본 논문에서는 한국과학기술원 멀티미디어 및 데이타베이스 연구실에서 개발하고 있는 오디세우스 객체관계형 DBMS를 위한 외부 조인을 설계하고 구현한다. 본 논문에서는 거의 모든 DBMS에서 제공 되는 가장 기본적인 조인 방법인 중첩 루프 조인 알고리즘을 확장하여 외부 조인 연산을 구현한다. 그리고, 외부 조인이 포함된 질의를 최적화하기 위해 조인 연산의 결과를 임시 릴레이션에 저장하는 대신에 다음 조인 연산의 입력으로 파이프라이닝시키는 것을 최대화 하는 방법을 제안한다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.11
• /
• pp.881-889
• /
• 2000

프로그램의 실행시간은 캐쉬메모리의 효율적 사용과 밀접한 관계가 있다. 특히 간섭 실패는 프로그램의 성능에 큰 영향을 미치지만 나타나는 형태가 불규칙적이므로 예측하기가 매우 어렵다. 본 논문에서는 직접 사상 캐쉬전략을 사용한 완전 중첩 루프 내 배열의 캐쉬 실패율(cache miss ratio)을 구하는 분석적 모델을 제시한다. 논문에서 제시한 모델을 임의의 캐쉬 위치에 각 배열이 접근한 시간을 기반으로 다음주기에서 캐쉬 실패의 발생 여부를 예측하는데, 간섭으로 발생한 캐쉬 실패 개수에 대해 기존에 제시된 모델보다 더 빠르고 정확한 예측이 가능하다. 특히, 한문장의 수행시간 예측시간은 배열의 크기와 독립적이기 때문에, 전체 프로그램의 수행시간 예측은 배열의 크기 및 문장의 반복 회수 배만큼 빠른 결과를 보여준다. 본 모델은 프로그램의 성능예측 뿐만 아니라 데이터 지역성의 최적화, 캐쉬 구성, 스케쥴링 등에서도 이용 가능하다.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.10
• /
• pp.1-10
• /
• 1994

In most application programs, loops usually contain most of the computation in a program and are the most improtant source of parallelism. When loops are executed on multiprocessors, the cross iteration data dependences need to be enforced by synchronization between processors. In this paper, we propose a new synchronization scheme(Free/Hold) for reducing overgeads occured by synchronization variables in data oriented scheme and delay of time occured by synchronization instruction in statement oriented scheme. The Free/Hold mechanism enforces the correct execution order by inserting synchronization instruction between each instance with data dependence relationship using the RD(Real dependence Distance). We also present an algorithm for removing unnecessary dependences in one-to-many dependences.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.34C no.6
• /
• pp.1-10
• /
• 1997

By considering low power design at higher levels of abstraction rather than at lower levels of abstraction, we can apply various transformation techniques to a system design with wider view and obtain much more effective power reduction with less cost and effort. In this paper, a transformation technique, called power - conscious loop folding is proposed for high level synthesis of a low power system.Our work is focused on reducing the power consumed by functional units in adata path dominated circuit through the decrease of switching activity. Te transformation algorithm has been implemented and integrated into HYPER, a high level synthesis system for experiments. In our experiments, we could achieve a pwoer reduction of up to 50% for data path dominated circuits.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.13 no.5
• /
• pp.775-784
• /
• 2010

In this paper we propose a design method of X band frequency hopping synthesizer in terms of phase noise and settling time with DDS driven PLL architecture, which has the advantages of high frequency resolution, fast settling time and small size. In addition, a noble method is proposed to remove the synthesizer output spurious signals due to superposition effect of DDS. The spurious signal which depend on its normalized frequency of DDS, can be dominant if they occur within the PLL loop bandwidth. We verify that the sources of that spurious signals are quasi-amplitude modulation and superposition effect, and suggest that such signals can be eliminated by intentionally creating frequency errors in the developed synthesizer.

• The Journal of the Korea Contents Association
• /
• v.7 no.12
• /
• pp.262-272
• /
• 2007

SAD (Sum of Absolute Difference) algorithm is the most frequently executing routine in motion estimation, which is the most demanding process in motion picture encoding. To enhance the performance of motion picture encoding on a VLIW processor, an optimal implementation of SAD algorithm on VLIW processor should be accomplished. In this paper, we propose an implementation of optimal scheduling of SAD algorithm with conditional branch on a VLIW-based high performance DSP. We first transform the nested loop with conditional branch of SAD algorithm into a single loop with conditional branch which has a large enough loop body to utilize fully the ILP capability of VLIW DSP and has a conditional branch to make the escape from loop to be achieved as soon as possible. And then we apply a modulo scheduling technique to the transformed single loop. We test the proposed implementation on TMS320C6713, and analyze the code size and performance with respect to processing time. Through experiments, it is shown that the SAD implementation proposed in this paper has small code size appropriate for embedded applications, and the H.263 encoder with the proposed SAD implementation performs better than other H.263 encoder with other SAD implementations.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.4
• /
• pp.653-660
• /
• 2001

In this paper, we designed 2303.15MHz frequency synthesizer for the purpose of the phase noise prediction. For the modeling of phase noise generated in the designed system through introducing the noise-modeling method suggested by Lascari we analyzed a variation of phase noise as according as that of offset frequency. Especially, for the third-order system of the PLL among some kinds of phase noise generated from VCO we analyzed the aspect of 1/f-noise appearing troubles in the low frequency band. Since it is difficult to analyze mathematically 1/f-noise in the third-order system of the PLL, introducing the concept of pseudo-damping factor has made an ease of the access of the 1/f-noise variance. we showed a numerical formula of 1/f-noise variance in the third-order system of the PLL which is compared with that of 1/f-noise variance in the second-order system of the PLL. As a result, In case of txco we found the reduce rapidly along the offset frequency after passed through that phase-noise was -160dBc/Hz before passed through a loop at 10kHz offset frequency and -162.6705dBc/kHz after passed through the loop, -180dBc/Hz at 100kHz offset frequency and -560dBc/kHz after passed through the loop. We can notice that the variance of third-order system more occurs (or the variance of second-order system in connection with noise bandwidth and variance factor of second-order and third-order system.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.10 no.6
• /
• pp.917-926
• /
• 1999

In this paper, the recording device for measuring the magnetic field waveforms associated with the cloud lightning discharges was made by the loop-type magnetic field sensor, and the measuring system of a distance between lightning stroke point and observatory site was proposed. The parameters including the time interval of superimposed pulses, zero-to-zero crossing time and the percentage depth of the dip for the positive and the negative magnetic field waveforms were statistically analyzed as a function of the distance between the lightning discharge point and the observatory site. The results could be summarized as follows; Several superimposed pulses were observed at the initial front part of the magnetic field waveform produced by cloud lightning discharges, and the mean time interval between superimposed pulses was about $4\mu\textrm{s}$. Also, the bipolar characteristic appeared significantly and the percentage depth of the dip was about 57~65%. It was known that the zero-to-zero crossing time of the electromagnetic fields is gradually decreased as a distance between lightning discharge point and observatory site is increased.

• Journal of Korea Spatial Information System Society
• /
• v.9 no.2
• /
• pp.25-34
• /
• 2007

Existing approaches that select an order for the join of three or more data streams have always used the simple heuristics. For their disadvantage - only one factor is considered and that is join selectivity or arrival rate, these methods lead to poor performance and inefficiency In some applications. The graph-based sliding window multi -join algorithm with optimal join sequence is proposed in this paper. In this method, sliding window join graph is set up primarily, in which a vertex represents a join operator and an edge indicates the join relationship among sliding windows, also the vertex weight and the edge weight represent the cost of join and the reciprocity of join operators respectively. Then the optimal join order can be found in the graph by using improved MVP algorithm. The final result can be produced by executing the join plan with the nested loop join procedure, The advantages of our algorithm are proved by the performance comparison with existing join algorithms.

• The korean journal of orthodontics
• /
• v.35 no.5 s.112
• /
• pp.371-380
• /
• 2005

The purpose of this experimental study was to evaluate aㅜd compare maxillary arch dimensional and positional changes between first and second premolar extraction groups. The Calorific Machine was used to illustrate tooth movement in three dimensions. The experimental teeth except the first or second premolars were embedded in artifical alveolar bone. The extraction space was closed using arch wires with bull loops into which 15 degree gable bends were placed. Before and after space closure, radiographs were taken in the sagittal and occlusal directions using occlusal films. The results showed greater mean maxillary incisor retraction and less anchorage loss in the maxillary first premolar extraction group than in the maxillary second premolar extraction group. Mesiopalatal rotation of anchor teeth was greater after extraction of a maxillary second premolar than a maxillary first premolar (P<.001).