This reference flow provides a unified co-design and analysis solution, enabled by Synopsys 3DIC Compiler to accelerate exploration and development of multi-die designs at all stages from silicon to systems. In addition, Synopsys 3DSO.ai is natively integrated with Synopsys 3DIC Compiler, enabling optimisation for signal, power and thermal integrity delivering productivity gains and maximum system performance.
"As bandwidth demands soar to new heights, companies are turning to multi-die designs at an accelerated pace to achieve greater levels of processing power and performance for their AI and high-performance computing applications," said Sanjay Bali, vice president of strategy and product management for the Synopsys EDA Group. "Our long-standing and deep collaboration with Intel Foundry, resulting in a production-ready AI-driven multi-die reference flow for EMIB technology, provides our mutual customers with a comprehensive solution that helps them develop their billion- to trillion-transistor multi-die systems."
Synopsys offers a comprehensive and scalable multi-die solution for fast heterogeneous integration that enables early architecture exploration, rapid software development and system validation, efficient die-package co-design, robust die-to-die connectivity, and improved manufacturing and reliability.
Adopted by multiple leading customers, Synopsys 3DIC Compiler, a key component of the multi-die solution, is integrated with Ansys RedHawk-SC Electrothermal multiphysics technology, to address the power and thermal signoff critical for 2.5D/3D multi-die designs. In addition, the solution maximizes system performance and quality of results at a rapid pace with Synopsys 3DSO.ai, an autonomous AI-driven optimisation engine for 2.5D and 3D multi-die designs.
Synopsys is developing IP for Intel Foundry process technologies, providing the interconnects needed to build multi-die packages with reduced integration risk and accelerated time-to-market. The combination of Synopsys IP and Synopsys 3DIC Compiler can enable up to 30% reduction in effort and 15% improvement in quality of results (as measured by margin) compared to traditional manual flows by automating routing, interposer studies, and signal integrity analysis.
