GBT Commences Research Project to Enhance Nanometer ICs Physical Verification Process


Reading time ( words)

GBT Technologies Inc., started a research project, internal name VeriSpeed, to develop new system and methods to enhance nanometer Integrated Circuits physical verification process.

A typical microchip design process involves few types of design’s verifications and validation. A design verification is the end-phase process to ensure that the integrated circuit works as planned. This process typically consumes vast amount of the total time spent on the microchip design. As microchips become more complex and advanced they include many more transistors and functions, all of which must be carefully tested and verified. A typical IC’s verification stages are functional design verification, which tests and validates the IC’s functionalities, physical design and verification, which checks the IC layout geometrical and electrical rules, packaging, and manufacturing tests.

During the physical design process a computer program is processing the microchip database, which is a huge size data. Typical processing of an advanced nanometer IC data, preparing it for physical verification, is a major time-consuming process. During this process the IC’s data is read and the program learns about the geometric shapes, which is called the mask layout. A set of mathematical algorithms then process this data to verify that all geometrical, electrical and DFM (Design For manufacturing) rules are met. The database engine is a key factor to the entire process performance.

VeriSpeed research project introduces a new approach to this essential element in this process, the IC’s data reading. By applying intelligent algorithms, the program will read the huge data according to a specific algorithm flow substantially speeding the entire verification process of nanometer and other Integrated Circuit’s layout Designs thus significantly impacting the global project’s design time cycle and manufacturing efficiency. The project aims to accelerate the entire IC’s verification process to enable reduction of total Integrated Circuit design time. This new database engine approach is targeted to work with GBT’s multiplanar methods as presented in the recent granted patent. 

“We are seeking to introduce an intelligent key factor within the IC physical design and validation process by addressing probably the most significant part of it, the database processing. An efficient database engine is a key factor for speed and high performance of the entire physical verification process including its three main aspects. The DRC, Design Rules Check – Geometrical Rules, LVS - Layout Versus Schematic, the electrical rules, and DFM - Design for Manufacturing rules.” stated Danny Rittman the Company’s CTO 

Based on advanced mathematical algorithm and deep learning the project will explore breakthrough approach and method for a new database engine that rapidly access to any specific area of the Integrated Circuit data. It is contemplated that the database engine will perform an initial analysis of the IC data to determine area’s priority for processing. The data then will be processed according to their density and complexity levels. The layout regions will be then processed in a fully hierarchical manner according to their internal indexing and priority weight.

In addition, unlike typical existing practice, it is expected that the proprietary mathematical algorithm will verify geometrical shapes in specific innovative method by split-and-merge segments to achieve higher speed. A neural network-based algorithm will validate the heuristic’s results to ensure on-track process. One of the main advantages of this system will be the early data process as the design moves forward. As the IC designer starts his/hers block design the system will learn the block’s data as the progresses. The database will be built incrementally over time and further shortened the final block physical verification process. The system will be working with GBT’s recent multiplanar 3D microchip methods as described in its patent.

Share

Print


Suggested Items

Kirigami Inspires New Method for Wearable Sensors

10/22/2019 | University of Illinois
As wearable sensors become more prevalent, the need for a material resistant to damage from the stress and strains of the human body’s natural movement becomes ever more crucial. To that end, researchers at the University of Illinois at Urbana-Champaign have developed a method of adopting kirigami architectures to help materials become more strain tolerant and more adaptable to movement.

Worldwide Semiconductor Equipment Billings at $13.3 Billion in 2Q19; Down 20%

09/12/2019 | SEMI
Worldwide semiconductor manufacturing equipment billings reached $13.3 billion in the second quarter of 2019, down 20% from the same quarter of 2018 and 3% from than the previous quarter.

Designing Chips for Real Time Machine Learning

04/01/2019 | DARPA
DARPA’s Real Time Machine Learning (RTML) program seeks to reduce the design costs associated with developing ASICs tailored for emerging ML applications by developing a means of automatically generating novel chip designs based on ML frameworks.



Copyright © 2021 I-Connect007. All rights reserved.