Carnegie Mellon University

Gradient Design Automation Inc.

Xigmix, Inc.

Larry Pileggi

Rajit Chandra

Peng Li

Mehdi Asheghi

Xin Liang Li

Herman Henry Schmit

Franz Franchetti

Fazle Sadi

Disclosed herein is a novel multi-way merge network, referred to herein as a Hybrid Comparison Look Ahead Merge (HCLAM), which incurs significantly less resource consumption as scaled to handle larger problems. In addition, a parallelization scheme is disclosed, referred to herein as Parallelization by Radix Pre-sorter (PRaP), which enables an increase in streaming throughput of the merge network. Furthermore, high performance reduction scheme is disclosed to achieve full throughput.

High performance merge sort with scalable parallelization and full-throughput reduction

A method and apparatus for modeling and thermal analysis of semiconductor chip designs is provided. One embodiment of a novel method for performing thermal testing of a semiconductor chip design includes calculating full-chip temperatures over the semiconductor chip design (e.g., to identify steep thermal gradients) and modeling the full-chip temperatures in accordance with a geometric multi-grid technique. The geometric multi-grid technique is tailored to determine temperatures within the semiconductor chip design based at least in part on the physical attributes or geometry of the design.

Method and apparatus for thermal modeling and analysis of semiconductor chip designs

A computer implemented method of performing projection based polynomial fitting. The method includes generating a plurality of sampling points as a function of variables. The method also includes forming a polynomial model template representative of the plurality of sampling points. According to embodiments of the present invention, the polynomial model template comprises at least one polynomial coefficient. The method further includes forming a low-rank matrix to approximate the polynomial coefficient.

Statistical optimization and design method for analog and digital circuits

A method is comprised of translating a bit stream defining the state of switches of an FPGA into a set of via geometries, or generating the set of via geometries directly from a physical design system. The via geometries are used to produce at least one via mask. The via mask is then used in a manufacturing process to customize an array of fixed and/or programmable logic blocks.

Growing community of inventors

Pittsburgh, PA, United States of America

Larry Pileggi