• Journal of Chest Surgery
• /
• v.33 no.9
• /
• pp.697-706
• /
• 2000

Background: Isolated lung perfusion(ILP) was developed as a new treatment approach to non-resectable primary or metastatic lung cancer, because of its ability to reduce systemic toxicity while delivering high-dose chemotherapeutic agents to the target organs. This research was planned to evaluate the direct toxic effect of high-dose cisplatin to the lung tissue during isolated lung perfusion. Material and Method: Fifteen mongrel dogs were divided in the perfusate for 40 minutes. The second group was composed of 5 mongrel dogs which underwent ILP with cisplatin 2.5 mg/Kg added to the perfusate for 30 minutes and 10 minutes with washing solution without cisplatin. The third group underwent the same procedure as the second group except cisplatin 5.0 mg/Kg in the perfusate. Activities of serum angiotensin converting enzyme(ACE), tumor necrosis factor-$\alpha$(TNF-$\alpha$), and concentration of serum lactate dehydrogenase(LDH) and blood urea nitrogen/creatinine (BUN/Cr) were analyzed in each groups at the time of pre-perfusion, 1 hour, 1 day, 1 week, and 2 weeks after ILP. Result: Serum ACE activities before and 1 hour, 1 day, 1 week, and 2 weeks after ILP in control group were 45.1$\pm$6.3, 44.6$\pm$9.3, 46.7$\pm$9.5, 50.8$\pm$9.1, 46.1$\pm$4.3 U/L. Those in cisplatin 2.5 and 5.0 mg/Kg groups were 49.4$\pm$12.6, 39.0$\pm$8.6, 42.3$\pm$15.9, 50.0$\pm$2.6, 53.8$\pm$8.3 and 55.5$\pm$12.3, 47.0$\pm$6.3, 45.1$\pm$6.9, 74.8$\pm$19.5, 60.2$\pm$12.0 U/L, respectively. Serum TNF-$\alpha$ activities in each group before and after ILP were 5.0$\pm$1.5 / 7.7$\pm$2.2 / 6.6$\pm$2.5 / 4.3$\pm$1.3 / 5.2$\pm$1.1(control), 8.7$\pm$1.6 / 9.9$\pm$2.2 / 7.9$\pm$1.5 / 6.3$\pm$2.2 / 7.4$\pm$2.4 (cisplatin 2.5 mg/Kg), and 6.9$\pm$0.7 / 8.9$\pm$3.4 / 7.9$\pm$4.0 / 3.3$\pm$0.9 / 5.8$\pm$1.3 pg/ml(cisplatin 5.0 mg/Kg). Mean LDH levels of each group were 225.7 / 271.3 / 328.9 / 350.8 / 255.7(control), 235.7 / 265.7 / 336.0 / 379.5 / 299.2 (cisplatin 2.5 mg/Kg), and 259.6 / 285.2 / 340.6 / 433.4 / 292.4 IU/L(cisplatin 5.0 mg/Kg). So there was no significant difference in serum ACE, TNF-$\alpha$, and LDH activity changes after ILP between the 3 groups. And, there was no significant changes in BUN/Cr in each groups, which was independent of ILP and perfused concentration of cisplatin. In addition, all dogs survived the ILP and there was no significant evidence of pulmonary vascular injury after 2 weeks of ILP with cisplatin. Conclusion: There was no harmful effect of cisplatin to the lund tissue of the mongrel dog up to 5.0 mg/Kg in perfusate. Therefore, it is perceived to be safe and effective to deliver high-dose cisplatin to the lung without pulmonary toxicity and renal damage with ILP.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.21-23
• /
• 1998

본 논문에서 ILP (Instruction Level Parallelism)의 성능향상을 위하여 데이터 값들을 미리 예측하여 병렬로 이슈(issue)하고 수행하는 기존의 데이터 값 예측기(data value predictor)를 비교 분석하여 각 예측기의 예측율을 측정하고, 2-단계 데이터 값 예측기(Two-Level Data Value Predictor)와 혼합형 데이터 값 예측기(Hydrid Data Value Predictor)에서 발생되는 aiasing 을 측정하기 위해 수정된 데이터 값 예측기를 사용하여 측정한 결과 aliasing은 50% 감소하였지만 예측율에는 영향을 미치지 못함과 데이터 값 예측기의 예측율을 측정한 결과 혼합형 데이터 값 예측기의 예측율이 2-단계 데이터 값 예측기와 스트라이드 데이터 값 예측기(Stride Data Value Predictor)에서 평균 5.7%, 최근 값 예측기(Last Data Value Predictor)보다는 평균 38%의 예측 정확도가 높음을 입증하였다.

• Journal of KIISE:Software and Applications
• /
• v.28 no.2
• /
• pp.201-209
• /
• 2001

고성능 마이크로 프로세서들은 성능 향상을 위해 ILP를 지원한다. 병렬성을 극대화시키기 위해서는 많은 성능 저해 요인들을 제거해야 한다. 최근에는 컴파일러의 역할을 증대시켜 이러한 요인들을 줄이기 위한 노력들이 활발히 진행되고 있다. 본 논문에서는 성능 저해 요인인 조건 분기 처리를 위하여 조건 실행과 레지스터 할당을 결합함으로써 메모리로의 대피를 최소화하고 병렬성을 향상시킬 수 있는 개선된 레지스터 할당 알고리즘을 제안한다. 제안한 방법을 적용하여 실험한 결과 간섭 그래프의 에지수가 4.47% 감소되었고 그 결과 요구되는 대피 변수의 수도 21.35% 감소되었다. 그리고 기존의 방법에 비해 19.38%의 성능 향상 결과를 얻었다. 결국 본 레지스터 할당 기법은 조건 실행을 통해 조건 분기 명령을 제거하여 기본 블록 내의 명령어 수를 증가시켜 병렬처리의 기회를 증진시키고 조건 분석을 통해 간섭 그래프의 불필요한 에너지를 제거시켜 보다 효율적인 레지스터 할당을 실현함으로써 제안한 방법의 타당성을 검증하였다.

• Journal of Internet Computing and Services
• /
• v.2 no.1
• /
• pp.23-30
• /
• 2001

ILP processors require an accurate branch prediction scheme to achieve higher performance. Two-Level branch predictor has been known to achieve high prediction accuracy. But, when a branch accesses a PHT entry that was, previously updated by other branch, Two-level predictor may cause interferences. Negative interferences among all interferences have a negative effect on performance, since they can cause branch mispredictions. Agree predictor achieve high prediction accuracy by converting negative interferences to positive interferences by adding bias bits to BTB, but negative interferences may occur when bias bit is set incorrectly. This paper presents a new dynamic branch predictor which reduces negative interferences. In the proposed predictor, we attach hit bits to entries in BTB to change bias bit dynamically during the execution time, h a result the proposed scheme improve the accuracy of prediction by reducing negative Interferences effectively, To illustrate the effect of the proposed scheme, we evaluate the performance of this scheme using SPEC92int benchmarks, The results show that the proposed scheme can outperform traditional branch predictors.

• KNOM Review
• /
• v.23 no.1
• /
• pp.34-42
• /
• 2020

Network Function Virtualization (NFV) environment can deal with dynamic changes in traffic status with appropriate deployment and scaling of Virtualized Network Function (VNF). However, determining and applying the optimal VNF deployment is a complicated and difficult task. In particular, it is necessary to predict the situation at a future point because it takes for the process to be applied and the deployment decision to the actual NFV environment. In this paper, we randomly generate service requests in Multiaccess Edge Computing (MEC) topology, then obtain training data for machine learning model from an Integer Linear Programming (ILP) solution. We use the simulation data to train the machine learning model which predicts the optimal VNF deployment in a predefined future point. The prediction model shows the accuracy over 90% compared to the ILP solution in a 5-minute future time point.

• Journal of KIISE:Software and Applications
• /
• v.26 no.5
• /
• pp.692-701
• /
• 1999

수퍼스칼라(superscalar)나 VLIW 와 같은 명령어 수준 병렬화(ILP) 프로세서의 성능을 극대화하는 과감한 명령어 스케쥴링은 소프트웨어 파이프라이닝과같은 스케쥴링 과정을 거치면서 일반적인 복사 명령어 제거 기법으로 없앨 수 없는 서로 간섭하는 복사 명령을 많이 만들어내는데 루프 내부에 생성된 이러한 복사명령은 적절한 루프 펼침을 수행하여 간섭관계를 없앰으로서 제거할 수 있다. 본 논문에서는 이와 같이 루프 펼침이 수행된 루프 내부의 복사명령을 제거하는 기법으로 그래프 컬러링 상에 구현한 낙관적 융합기법을 제안한다. 그래프 컬러링에서의 융합기법은 간선의 개수가 많은 노드를 만들어 낼수 있으므로 채색성에 부정적인 영향을 주는 것으로 알려져 왔으나 본 기법에서는 융합되는 노드에 동시에 간섭하는 노드의 간선의 수가 줄어드는 긍정적인 영향을 최대한 이용하여 채색성을 높이고 융합된 노드에 대한 실제 버림(spill)이 일어나는 경우 유효 범위 분절(live range splitting)을 통하여 버림의 부담을 최대한 줄이도록 하였으며 이를 VLIW 스케쥴링 된 SPEC 정수벤치마크 루프내부의 복사 명령 제거에 적용한 결과 제거 가능한 복사 명령의 99%를 제거하면서도 버림명령은 다른 융합 기법과 비교하여 가장 적게 발생하는 우수한 결과를 얻을수 있었다.

• Journal of Korea Game Society
• /
• v.8 no.2
• /
• pp.69-76
• /
• 2008

In this paper, we present the architecture of graphics processor and its instruction format for the mobile device. In addition, we introduce tile shader data structure for the on/off-line compilation based on the OpenGL ES 2.0 and a new optimization method based on the ILP(Instruction-Level Parallelism). This paper shows where a processor with the sane core clock is being used, the shader instruction resulted from the compile structure and method in this paper is approximately 1.5 to 2 times faster than a code based on the single instruction.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.11B
• /
• pp.663-672
• /
• 2007

This paper proposes an algorithm for design of a multiple ring network that can be applied to an arbitrary topology in the WDM optical transport networks. Especially, we suggest an integer linear program(ILP) formulation with the objective of minimizing the total cost of nodes based on the reconfigurable OADM(ROADM) architecture and propose a heuristic algorithm to reduce computational complexity of the exact problem formulation. To evaluate the performance and validity of this scheme, we applied it to an NSFnet topology assuming the dynamic or static traffic condition and compared it with a mesh methodology about capacity of connection requests and the total cost of the nodes.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.451-452
• /
• 2008

In this paper we describe a multi-module, multi-port memory design procedure that satisfies area and/or energy constraints. Our procedure uses ILP models to determine (a) the memory configuration with minimum area, given the energy bound, (b) the memory configuration with minimum energy, given the area bound. If we have a margin in time constraint, we break up conflict edges and expend the search space of ILP. This method effectively reduces area and power of the designed results.

• Journal of the Korea Society of Computer and Information
• /
• v.7 no.2
• /
• pp.40-45
• /
• 2002

In this paper, we deals with scheduling problem for data-path synthesis in high-level synthesis. We proposed a new method to optimize interconnection cost for operator allocation. Especially, we focus to optimize for buses to transfer data channel and register to restore operation result. Also we take account ILP formulation in order to get optimal scheduling result. To verify the effectiveness of this study, we select 5th Digital Wave Filter. The experimental result show that purposed method was better then the general methods.