Our recent work, led by Ph.D. student, Han Zhou, to optimize the on-chip power grid networks considering the electromigration effects using machine learning based approach has been accepted by TCAD. This work represents our new approach to addressing the challenging EM failure and aging issues in the nanometer VLSI chip design. Our study shows that...
Most of senior PhD students in VSCLAB have received internship job offers for summer 2022. Wentian will work for Synopysis for machine learning related projects. Jinwei Zhang will work for Nvidia. Shuyuan will work for PrimeShield group in Synopsys. Subed Lamichhane will work in Synopsys and Yibo Liu will also work in the Synopsys for...
Ph.D. student Wentian Jin received the Dissertation Year Program (DYP) fellowship award from UCR graduate division. This is very competitive fellowship awards from the UCR graduate division for the last-year Ph.D. students. Also first-year Ph.D. student Chinmay Raje received DAC Young Fellow award to attend DAC 2022. Congratulation on both of them!
Prof. Sheldon Tan has been selected to serve as the IEEE CEDA Distinguished Lecturer for 2022-2023. Under this DL program, Prof. Tan will delivers a few lectures over the time 2022-2023 to the EDA community to disseminate the latest research works developed in the VSCLAB. The mission of the CEDA Distinguished Lecturer Program (DLP) is...
Dr. Liang Chen's work on the electrothermal modeling and optimal design has been accepted by IEEE Transaction on TCAD. This work develops a novel compact electrothermal devices (TEC) based on the first principles. Based on this model, the optimal voltages can be derived analytically for a given TEC devices for maximum cooling capacity. L. Chen...
Three papers have been accepted by ASPDAC 2022 as below. The first paper is about new graph convolution network based analysis techniques for emerging chipset design. The second paper is about new fast EM analysis technique consider Joule heating effects. The third paper is about new approximate multiplier design based logarithmic multipliers. Congratulation on the...
NSF funded the research of VSCLAB for for machine learning based thermal monitoring and run-time thermal management for manycore processors and chiplet designs. Prof. Sheldon Tan is the single PI for this research work with 500K budget for three years. Today’s high-performance processors, and even emerging mobile platforms, are more thermally constrained than ever before...
The research work for performing full-chip power and thermal map estimation for commercial multi-core microprocessor under realistic heat sink cooling has been accepted by IEEE Transaction on TCAD, the top journal in the EDA field. This work, leaded by Ph.D. student, Jinwei Zhang, is the first work to accurate estimate the full-chip power and thermal...
The work by Ph.D. student Sherif Sadiqbatcha on real-time full-chip thermal estimation based on the machine learning method for commercial multi-core CPU has been accepted by IEEE Transaction on Computer, the top journal for computer architecture. The is significant work from VSCLAB for using machine learning to generate the full-chip map directly from the tasks...
Most of senior PhD students in VSCLAB have received internship job offers for summer 2021. Sheriff Sadiqbatcha will continue work for Intel Corp. Wentian will work for Synopysis for machine learning related projects. Jinwei Zhang will work for PrimeTime in Synopsys, Shuyuan will work for PrimeShield group in Synopsys.
The PhD students and Dr. Sheldon Tan got together for the first time on March 26, 2021 for a hiking in the park in Central Ave in Riverside and dinner later in the Riverside plaza for the first time since the pandemic. It is really a joyful to meet most everyone in person for the...
Ph.D. student, Shaoyi Peng successfully defensed her thesis titled "Modeling and Simulation methods for VLSI Interconnect Reliability Focusing on TDDB". Shaoyi joined the VSCLAB in 2016 and he worked on the VLSI reliability modeling and GPU based parallel simulation. Shaoyi will join the Cerebras System, an AI chip startup as a senior engineer after graduation...
Ph.D. student, Han Zhou successfully defensed her thesis titled "Electromigration-Aware On-Chip Power Grid Design and Optimization". Han joined the VSCLAB in 2016 and she worked on the EM power grid optimization. Han will join Synopsys's IC compiler group as a senior engineer after graduation. We wish her the best for her future endeavors.
Two DAC 2021 paper from VSCLAB have been accepted. The first paper, led by PhD student Shyuan Yu, proposes a new approximate multiple using improved stochastic computing (called counter-based SC) techniques and showed significant advantages over existing state of the art approximate adders on DNN network applications. Congratulations on Shuyuan. COSAIM: Counter-based Stochastic-behaving Approximate Integer...
Our recent work based on the physics-constrained deep learning framework for electrostatics analysis was published in DATE 2021. In this work, we developed a label free self-learning method to solve differential equations for electrostatic analysis, which is important for reliability and capacitance extraction method. W. Jin, S. Peng, and S. X.-D. Tan, “Data-driven electrostatics analysis...
Two papers were published in MLCAD, the new ACM/IEEE conference focusing on the machine learning techniques for design automation techniques. The first paper by Ph.D. student Shaoyi Peng developed an auto-encoder/decoder like DNN networks for analysis TDDB reliability effects. The second paper by Ph.D. student Jinwei Zhang developed an efficient reinforcement learning based hotspot-aware task...
Three papers from our lab have been published recently. The first paper/work exploits the differential natural of DNN networks to significantly speedup the gradient based optimization method for power grid networks subject to electromigration constraints. This work was led by Ph.D. student Han Zhou and it also was nominated as the Best Paper Award. H...
Our recent paper published in ICCD 2021 has been nominated as the Best Paper Award. Ph.D. student Wentian Jin developed new GAN (Generative Adversary Network) based analysis method for solving the partial differential equations for electromigration. W. Jin, S. Sadiqbatcha, Z. Sun, H. Zhou and S. X.-D. Tan, “EM-GAN: Data-driven fast stress analysis for multi-segment...
Our recent work for using stochastic computing for dynamic reliability and power management has been accepted by the top embedded conference, CASES'20 (ESWEEK 2020). This is the first work to explore the dynamic reliability management using the improved stochastic computing method. The work is led by Ph.D. student Shuyuan Yu. S. Yu, H. Zhou, H...
Elsevier just released the new CiteScore and Impact Factor for the Integration, the VLSI Journal, the CiteScore (CS) is 2.6, Impact Factor (IF) is 1.214. This is a good improvement from 2.2 for CS and 1.15 for IF from 2018. Prof. Tan was the Editor in Chief for Integration since 2016. CiteScore of a journal...