A Collaborative Data Model For AI/ML In EDA


This work explores industry perspectives on: Machine Learning and IC Design Demand for Data Structure of a Data Model A Unified Data Model: Digital and Analog examples Definition and Characteristics of Derived Data for ML Applications Need for IP Protection Unique Requirements for Inferencing Models Key Analysis Domains Conclusions and Proposed Future Work Abstra... » read more

Mythic Case Study


Mythic, the provider of a unique AI compute platform, was designing an innovative intelligence processing unit (IPU) and found themselves in need of a small, power-efficient, yet programmable core to take care of specific supporting functions. As no off-the-shelf core would exactly match the needs and customization proved challenging, Mythic eventually opted for a complete solution by Codasip. ... » read more

Week In Review: Design, Low Power


Tools, Cloud, IP Valtrix Systems updated its STING design verification tool for RISC-V based CPU and SoC implementations. Version 1.9.0 adds support to verify recent changes to the RISC-V user and privilege specifications, including draft versions of the vector and bit manipulation standard extensions. Preliminary support for the draft version of the RISC-V hypervisor extension has also been a... » read more

Blog Review: Dec. 16


Arm's Benoit Labbe investigates why battery monitoring is so important for a low-power microcontroller and shows how it was implemented in the M0N0 MCU while drawing a fraction of a nW in typical conditions. Siemens EDA's Harry Foster takes a look at how much of their time FPGA design engineers spend on verification, and the tasks that keep verification engineers the busiest. Synopsys' Sc... » read more

Week In Review: Design, Low Power


RISC-V RISC-V International CEO Calista Redmond provided an update on the state of the community during the annual RISC-V Summit: “RISC-V has had an incredible year of growth and momentum. This year, our technical community has grown 66 percent to more than 2,300 individuals in our more than 50 technical and special interest groups. We’re seeing increased market momentum of RISC-V cores, S... » read more

Detecting Electrical Hazards Incurred By Inter-Voltage Domain Crossing In Custom SRAMs


Fast-growing markets, such as 5G, biotechnology, AI, and automotive, are driving a new wave of low-power semiconductor design requirements and, hence, more aggressive low-power management techniques are needed. Consequently, even large macros within a chip, such as SRAMs, now feature multiple voltage domains to limit power draw during light-sleep, deep-sleep, and shutdown-low-power modes. These... » read more

Blog Review: Dec. 9


Arm's Benoit Labbe digs into designing a power converter for Arm Research's ultra-low power M0N0 microcontroller, with a focus on optimal efficiency and leakage constraints. Mentor's Harry Foster tries to get a sense of how much effort is spent in verification of FPGAs by looking at the amount of time spent and number of engineers on a project. Cadence's Paul McLellan listens in as Odile ... » read more

Week In Review: Design, Low Power


Xilinx acquired the assets of Falcon Computing Solutions, a provider of high-level synthesis (HLS) compiler optimization technology for hardware acceleration of software applications. The acquisition will be integrated into the Xilinx Vitis Unified Software Platform to automate hardware-aware optimizations of C++ applications with minimal hardware expertise. “Our compiler provides a high degr... » read more

Blog Review: Dec. 2


Mentor's Harry Foster investigates the effectiveness of today’s FPGA verification processes in terms of nontrivial bug escapes into production as part of the 2020 Wilson Research Group Functional Verification Study. Synopsys' Chris Clark points to how integral sensors are to the modern vehicle and key design considerations for making them more effective, safe, and reliable. Cadence's Pa... » read more

Forward And Backward Compatibility In IC Designs


Future-proofing of designs is becoming more difficult due to the accelerating pace of innovation in architectures, end markets, and technologies such as AI and machine learning. Traditional approaches for maintaining market share and analyzing what should be in the next rev of a product are falling by the wayside. They are being replaced by best-guesses about market trends and a need to bala... » read more

← Older posts Newer posts →