In VLSI layout design, certain nets in a given net set are required to propagate their signals within a tolerable skew of delays. Though the delay of the signal on a wire is determined by a complex environment, it is hard to satisfy this requirement unless all the concerned nets are routed within a certain skew of length. In this paper, we propose L-equidistance routing, which routes the concerned nets with a prescribed length L. After a basic technique of L-equidistance routing of a single 1-sink net, an algorithm is presented for the channel routing of plural multi-sink nets. The key idea is in the symmetric-slant grid interconnect scheme by which the problem is reduced to a grid routing problem. In L-equidistance routing of a channel, the total length of a n-sink net is not unique for nX3: An algorithm based on dynamic programming to solve this minimization problem is presented. Then, L-equidistance switch-box routing is discussed based on the L-equidistance channel routing. Algorithms are explained on the Euclidean space. But it is shown that a straightforward transformation of the routes to those on the Manhattan grid is possible keeping the property of equidistance. The proposing channel routing algorithm was implemented and applied to random data to demonstrate their ability.

No abstract

No abstract

No abstract

No abstract

No abstract

The 3-dimensional (3D) ICs' increased module density exacerbates the thermal hot-spot problem: A larger module packed into a smaller footprint produces a higher maximum temperature. Because of the significant impact of thermal via on lowering the thermal resistance of the chip, an appropriate thermal via management can be hoped to alleviate the unfavorable thermal phenomena of 3D ICs. In this paper, based on a finite difference (FD) thermal mesh model, we firstly present an iterative thermal via planning (FD-TVP) algorithm for 3D standard cell layout, and then two methods are devised to realize the thermal via density's minimization during the chip's placement stage: (1) one is FD-TVP-Fast method, in which a thermal-aware Gravity placement algorithm is followed by the FD-TVP operation; (2) the other is FD-TVPPlacement, which is hoped to obtain a final placement with a good thermal characteristic by introducing FD-TVP to the Gravity placement process. Finally, the simulations on the IBM-PLACE benchmarks demonstrate our algorithms can achieve the maximum and average temperature objectives while minimizing the thermal via utilization. Moreover, the comparison between the results of FD-TVP-Fast and FD-TVP-Placement is also shown.

No abstract