Patrick H. Madden

Manhattan or non-Manhattan?: a study of alternative VLSI routing architectures

Circuit interconnect has become a substantial obstacle in the design of high performance systems. In this paper we explore a new routing paradigm that strikes at the root of the interconnect problem by reducing wire lengths directly. We present a non-Manhattan Steiner tree heuristic, obtaining wire length reductions of much as 17% on average, when compared to rectilinear topologies. Moreover, we present a graph-based interconnect optimization algorithm, called the GRATS-tree algorithm, which allows performance optimization beyond what can be obtained through wire length reduction alone. The two tree construction algorithms are integrated into a new global router that allows large scale non-Manhattan design. Although we consider circuit placements performed under rectilinear objectives, our global router can reduce maximum congestion levels by as much as 20%. In general we find that the non-Manhattan approach requires additional Steiner points and bends; realization of non-Manhattan routing structures requires additional vias. We observe that the increase in via cost is much less dramatic than might be expected; the benefits of wire length reduction may outweigh the additional via cost.

Routability-Driven Placement and White Space Allocation

We present a congestion-driven placement flow. First, we consider in the global placement stage the routing demand to replace cells in order to avoid congested regions. Then we allocate appropriate amounts of white space into different regions of the chip according to the congestion map. Finally, a detailed placer is applied to legalize placements while preserving the distributions of white space. Experimental results show that our placement flow can achieve the best routability with the shortest routed wirelength among all publicly available placement tools. Moreover, our white space allocation approach can significantly improve the routabilities of placements generated by other placement tools.

Improved global routing through congestion estimation

In this paper, we present a new method to improve global routing results. By using an amplified congestion estimate to influence a rip-up and reroute approach, we obtain substantial reductions in total congestion. In comparisons with recently published tool on publicly available benchmarks, our new router is roughly twice as fast, obtains 15.1% reductions in total wire length, and 65.2% reductions in the number of overcongested graph edges. A direct implementation of an old approach also performs well, indicating that some known techniques have been overlooked.

Routability-driven placement and white space allocation

We present a congestion-driven placement flow. First, we consider in the global placement stage the routing demand to replace cells in order to avoid congested regions. Then we allocate appropriate amounts of white space into different regions of the chip according to the congestion map. Finally, a detailed placer is applied to legalize placements while preserving the distributions of white space. Experimental results show that our placement flow can achieve the best routability with the shortest routed wirelength among all publicly available placement tools. Moreover, our white space allocation approach can significantly improve the routabilities of placements generated by other placement tools.

Improved cut sequences for partitioning based placement

Recursive partitioning based placement has a long history, but there has been little consensus on how cut sequences should be chosen. In this paper, we present a dynamic programming approach to cut sequence generation. If certain assumptions hold, these sequences are optimal. After study of these optimal sequences, we observe that an extremely simple method can be used to construct sequences that are near optimal. Using this method, our bisection based placement tool Feng Shui outperforms the previously presented Capo tool by 11% on a large benchmark. By integrating our cut sequence method into Capo, we are able to improve performance by 5%, bringing the results of Feng Shui and Capo closer together.

Recursive bisection placement: feng shui 5.0 implementation details

In this paper, we summarize circuit placement techniques and algorithms developed by the BLAC CAD research group; these have been integrated into our recursive bisection based placement tool feng shui. We also briefly describe current research interests.

Benchmarking for large-scale placement and beyond

Over the last five years, the large scale integrated circuit placement community achieved great strides in the understanding of placement problems, developed new high-performance algorithms, and achieved impressive empirical results. These advances have been supported by a nontrivial benchmarking infrastructure, and future achievements are set to draw on benchmarking as well. In this paper, we review motivations for benchmarking, especially for commercial electronic design automation, analyze available benchmarks, and point out major pitfalls in benchmarking. Our empirical data offers perhaps the first comprehensive evaluation of several leading large-scale placers on multiple benchmark families. We outline major outstanding problems and discuss the future of placement benchmarking. Furthermore, we attempt to extrapolate our experience to circuit layout tasks beyond placement.

Mixed block placement via fractional cut recursive bisection

Recursive bisection is a popular approach for large scale circuit placement problems, combining a high degree of scalability with good results. In this paper, we present a bisection-based approach for both standard cell and mixed block placement; in contrast to prior work, our horizontal cut lines are not restricted to row boundaries. This technique, which we refer to as a fractional cut, simplifies mixed block placement and also avoids a narrow region problem encountered in standard cell placement. Our implementation of these techniques in the placement tool Feng Shui 2.6 retains the speed and simplicity for which bisection is known, while making it competitive with leading methods on standard cell designs. On mixed block placement problems, we obtain substantial improvements over recently published work. Half perimeter wire lengths are reduced by 29% on average, compared to a flow based on Capo and Parquet; compared to mPG-ms, wire lengths are reduced by 26% on average.

Floorplan management: incremental placement for gate sizing and buffer insertion

Incremental physical design is an important methodology towards achieving design closure for high-performance large-scale circuits. Placement tools must accommodate incremental changes to the layout and netlist due to physical synthesis techniques without perturbing the original metrics. We present an incremental placement approach using floorplan sizing to manage the resources and demands of the whole chip region in order to accommodate the changes due to gate sizing and buffer insertion. The experimental results show that this approach can accommodate a wide range of incremental changes without a loss in wirelength and mutability. Most important, it also maintains the stability of a placement such that the convergence of physical synthesis iterations can be greatly enhanced.

Reporting of standard cell placement results

Very large scale integration (VLSI) fabrication technology has advanced rapidly, bringing with it a strong demand for faster and better design automation tools. Accurate reporting of results for placement approaches is crucial to the development of improved automation tools; unfortunately, publicly available placement benchmarks are outdated, and there are wide variations in their interpretation. In addition, the metrics considered by some academic research have questionable relevance to modern design. At best, poor benchmarks and differences in interpretation result in misunderstandings of the effectiveness of some approaches. At worst, they can motivate research in areas of very little promise, while other areas which have true potential are ignored. In this paper, we expand on work previously presented, describing current standard cell placement benchmarks and illustrating common differences in their interpretation. We also propose specific interpretation methods for traditional objectives, and discuss new metrics which should be considered in modern placement research. Our hope is that by presenting these issues clearly, we can enable more accurate evaluations of placement methods, and improve research efficiency.