Wilm E. Donath

Timing driven placement using complete path delays

The Timing Drive Placement (TDP) system balances wirability and timing constraints so that the final released design meets timing criteria. This is achieved by dynamically evaluating the timing of critical paths during placement. TDP is significant because convergence to a timed wirable solution early in the physical design cycle is achieved, or else it becomes apparent that logic changes are required.

Prediction of wiring space requirements for LSI

A stochastic model is developed for estimating wiring space requirements for one-dimensional layouts. This model uses as input the number of devices in the complex to be wired, the average length of a connection, and the average number of connections per device, to compute the probability of successfully wiring the devices as a function of the number of tracks provided. A heuristic approach is used to extend this model to the two-dimensional case, and tested against experimen-tal studies. Satisfactory agreement is found between a priori calculations of track requirements for the two-dimensional case against global wiring solutions for artificially generated problems, and for some layouts of actual logic complexes.

Complexity Theory and Design Automation

Complexity Theory is discussed and its relationship to Physical Design (i.e. Placement/Wiring) and Test Pattern Generation is shown and developed.

Transformational placement and synthesis

Novel methodology and algorithms to seamlessly integrate logic synthesis and physical placement through a transformational approach are presented. Contrary to most placement algorithms that minimize a global cost function based on an abstract representation of the design, we decomposed the placement function into a set of transforms and coupled them directly with incremental timing, noise, and/or power analyzers. This coupling results in a direct and more accurate feedback on optimizations for placement actions. These placement transforms are then integrated with traditional logic synthesis transforms leading to a converging set of optimizations based on the concurrent manipulation of boolean, electrical, as well as physical data. Experimental results indicate that the proposed approach creates an efficient converging design flow that eliminates placement and synthesis iteration. It results in timing improvements, and maintains other global placement measures such as wire congestion and wire length. The flexibility of the transformational approach allows us to easily add, extend and support more sophisticated algorithms that involve critical as well as non-critical regions and target a variety of metrics including noise, yield and manufacturability:.

Demonstrating the Scalability of a Molecular Dynamics Application on a Petaflops Computer

The IBM Blue Gene/C parallel computer aims to demonstrate the feasibility of a cellular architecture computer with millions of concurrent threads of execution. One of the major challenges in this project is showing that applications can successfully scale to this massive amount of parallelism. In this paper we demonstrate that the simulation of protein folding using classical molecular dynamics falls in this category. Starting from the sequential version of a well known molecular dynamics code, we developed a new parallel implementation that exploited the multiple levels of parallelism present in the Blue Gene/C cellular architecture. We performed both analytical and simulation studies of the behavior of this application when executed on a very large number of threads. As a result, we demonstrate that this class of applications can execute efficiently on a large cellular machine.

Pariser—Parr Calculations and Electro‐Optical Effects in Benzene

Pariser—Parr calculations were done for the singlet states of benzene using singly and doubly excited configurations. Good agreement is obtained with the experimental locations of the 1B2u, 1B1u, and 1E1u states as well as the oscillator strength of the 1A1g→1E1u transition. The 1E2g state is calculated to be 0.5 eV above the 1E1u state and its photodissociation can be deduced from its bond orders. A possible assignment of the 1A1g→1E2g transition is discussed. On the basis of this simple model, electro‐optical effects are small, such as the Stark shift for the 1E1u and 1B1u states.

Demonstrating the scalability of a molecular dynamics application on a Petaflop computer

The IBM Blue Gene project has endeavored into the development of a cellular architecture computer with millions of concurrent threads of execution. One of the major challenges of this project is demonstrating that applications can successfully exploit this massive amount of parallelism. Starting from the sequential version of a well known molecular dynamics code, we developed a new application that exploits the multiple levels of parallelism in the Blue Gene cellular architecture. We perform both analytical and simulation studies of the behavior of this application when executed on a very large number of threads. As a result, we demonstrate that this class of applications can execute efficiently on a large cellular machine.

Vibrational Interactions in Pariser—Parr Theory. I. Bandwidths and Transition Intensities

It is shown that bandwidths of some transitions can be calculated with good accuracy in the Pariser—Parr theory. For these calculations the critical point is a knowledge of the variation of the bond resonance integral β with respect to distance, which is given by κ=−dln(−β)/dR=2.0 A−1. Calculations are carried out for the bandwidths of the N–V transition of ethylene and the 1A1g→1E1u, 1E2g transitions of benzene and the intensities of the 1A1g→1B2u, 1B1u transitions of benzene. Satisfactory agreement with experiment is obtained; the neglect of electronic repulsion terms is demonstrated to have a relatively small effect. Selection rules are established for the strength of vibrational coupling in alternant hydrocarbons.

Vibrational Interactions in Pariser—Parr—Pople Theory. II. Ground‐ and Excited‐State Force Fields

Expressions are derived for the interaction between ground and excited states to give resonance contributions to the vibrational force field. Two contributions are found: The dominating term reflects largely the change in the bond resonance integral due to bond length variation with some contribution of the electrostatic terms; the other, a relatively small term, is generated by the polarization due to the field generated by a vibrational distortion. Calculations are made for the ground states of a number of aromatics; adequate agreement with experiment is found for benzene and naphthalene.The possible distortion of molecules and the Born—Oppenheimer approximation are investigated when the resonance interactions become strong. Application is made to the 1B1u and 3B1u states of benzene. Distortion and violation of the Born—Oppenheimer approximation is predicted to occur for the two B1u states of benzene.

Atomic Calculations. II. Polarizabilities and London Force Constants for F—, Ne, and Na+

Calculations are described for the polarizabilities and London force coefficients for F—, Ne, and Na+. The ground‐state wavefunctions are with and without correlation. The perturbing functions are singly and doubly excited configurations, which form a set of pseudostates; the polarizability is then calculated by second‐order perturbation theory. The various effects occurring here are considered.