• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.7
• /
• pp.21-28
• /
• 2017

Practical models for FPGA architectures which include performance- and/or density-enhancing components such as carry chains, wide function multiplexers, and memory/multiplier blocks are being applied to academic FPGA placement tools which used to rely on simple imaginary models. Previously the techniques such as pre-packing and multi-layer density analysis are proposed to remedy issues related to such practical models, and the wire length is effectively minimized during initial analytic placement. Since timing should be optimized rather than wire length, most previous work takes into account the timing constraints. However, instead of the initial analytic placement, the timing-driven techniques are mostly applied to subsequent steps such as placement legalization and iterative improvement. This paper incorporates the timing driven techniques, which check if the placement meets the timing constraints given in the standard SDC format, and minimize the detected violations, with the existing analytic placer which implements pre-packing and multi-layer density analysis. First of all, a static timing analyzer has been used to check the timing of the wire-length minimized placement results. In order to minimize the detected violations, a function to minimize the largest arrival time at end points is added to the objective function of the analytic placer. Since each clock has a different period, the function is proposed to be evaluated for each clock, and added to the objective function. Since this function can unnecessarily reduce the unviolated paths, a new function which calculates and minimizes the largest negative slack at end points is also proposed, and compared. Since the existing legalization which is non-timing driven is used before the timing analysis, any improvement on timing is entirely due to the functions added to the objective function. The experiments on twelve industrial examples show that the minimum arrival time function improves the worst negative slack by 15% on average whereas the minimum worst negative slack function improves the negative slacks by additional 6% on average.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.1_2
• /
• pp.92-104
• /
• 2006

The proposed method for a force directed global placement algorithm exploits and extends techniques from two leading placers, Kraftwerk (& KraftwerkNC) and Mongrel. It combines the strengths of KraftwerkNC, force directed global placer, and Mongrel's ripple move technique which resolves cell overlaps effectively The proposed technique uses the force spreading technique used in Kraftwerk to optimize the ripple movement. While it is resolving the cell overlap and optimizing wire length physical net constraints are considered for timing. The experimental results obtained by the proposed approach shows significant improvement on wire length as well as on timing.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10c
• /
• pp.375-379
• /
• 2007

고성능 VLSI 설계 시 배치 후를 포함한 전체적인 설계 과정이 완성되기 전까지는 물리적인 정확한 설계의 특성은 배치 단계에서는 알기 어렵다. 따라서 주어진 성능 (시간적 제약조건)을 만족하는, 즉, timing-driven placement (타이밍이 고려된 배치)는 1.0 미크론 이하의 초미세한 설계에서 중요하게 되었다. 타이밍을 고려한 배치는 초기 레이아웃 디자인 단계에서 타이밍 제약조건에 의해 디자인 반복을 줄인다. 하지만 대부분의 배치 단계의 디자인 모델은 배치단계에서 기하학적인 면을 고려하여 최대허용 지연시간 (Slack 이라고 부름)과 같은 물리적인 디자인 효과를 분석하기 어려운데 이것은 물리적으로 정확한 특성이 이 단계에서 알려지지 않기 때문에 당연한 결과이다. 본 논문에서는 기하적인 요소를 고려한 Slack의 재분배의 이점을 이용하여 허용 지연시간 처리의 혁신적인 방법을 제안한다. 제안된 접근법은 timing-closed 솔루션을 쉽게 찾는데 도움을 주고 이는 디자인을 반복하는 시간을 절약할 수 있게 한다.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.39 no.6
• /
• pp.34-43
• /
• 2002

This paper deals with a standard cell placement which is based on a quadratic programming. This paper proposes a new additional force to reduce the cell overlap and to get a uniform distribution of cells. The additional force is not concerned with interconnections between cells, but it is determined by the density of a placement area. In this paper, we modelled that the new additional force is a force which is caused by the dummy fixed cell. And it is used for the global placement. Proposed placement method is compared with TimberWolf v7.0 and Itools vl.4. Proposed placer achieved 7.5% average reduction in wirelength in non timing driven mode, 5.0% average reduction in wlrelength in timing driven mode compared to TimberWolf v7.0. And we got a comparable result to Itools vl.4.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.6
• /
• pp.354-360
• /
• 2002

Delay estimation must be simple and efficient, since millions or more delay calculations may be required during a timing-driven placement stage. We have developed a new Modified Elmore delay estimation method, which is significantly more accurate than the original Elmore delay by considering resistance shielding effects, but has the same order of complexity with that of Elmore delay. Experimental results show that the suggested technique can significantly reduce the error in estimated delay, from 31.6 ~ 145.2% to 2.5 ~ 22.7%.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.11a
• /
• pp.853-856
• /
• 2005

VLSI 설계에서 셀 배치단계는 칩의 성능에 영향을 미치는 중요한 단계로서, 셀 배치문제의 주요한 목적비용으로는 배선길이, 타이밍(timing) 그리고 배선밀집도 (routing congestion)가 있다. 기존 연구에서 배선길이를 줄이기 위한 많은 기법들이 소개되었으나 배선 밀집도를 추정하고 이를 어떻게 줄일 것인가에 대한 연구는 상대적으로 많이 되어있지 않다. 본 논문에서는 셀 배치후에, 주어진 배치를 바탕으로 배선밀집도를 예측하고 배선밀집도가 높은 지역을 국부적으로 해결하는 새로운 기법을 제안한다.