industry news
Subscribe Now

Graphcore leverages multiple Mentor technologies for its massive, second-generation AI platform

Graphcore has used a range of technologies from Mentor, a Siemens business, to successfully design and verify its latest M2000 platform based on the Graphcore Colossus™ GC200 Intelligence Processing Unit (IPU) processor. 

Integrating 59.4 billion transistors on a single 823sqmm die and manufactured on TSMC’s 7nm processes, Graphcore’s second-generation IPU is one of the most complex processors ever made. To tackle this design challenge, Graphcore leveraged multiple Mentor solutions to handle a range of critical tasks including circuit verification, PCB design, protocol verification, thermal analysis, design-for-test (DFT) and bring-up of the state-of-the-art AI processor.

“Mentor’s extremely comprehensive portfolio of world-class IC and PCB design automation tools delivered the capacity, innovation, IP and trusted flows we needed for a design of this magnitude,” said Phil Horsfield, vice president of Silicon at Graphcore. “The performance and capacity of Mentor’s solutions allowed us to optimize the entire electronics system and focus on our differentiating IP, which helped us deliver a truly unique and compelling end-product.”

To deliver its second-generation IPU, Graphcore used Mentor tools including: 

  • The industry-leading Calibre® platform, which includes Calibre nmDRC, the industry-leading physical verification solution, and Calibre nmLVS, the industry-leading circuit verification solution, both of which helped Graphcore support design intent for its second-generation device. Graphcore also used Calibre YieldEnhancer, featuring SmartFill, to control design planarity and help reduce turnaround time across multiple design iterations.
  • The Questa™ Verification IP solution for PCI Express 4.0, used to verify the Graphcore IPU’s PCIe Gen4, Ethernet and AMBA interconnect I/O busses. Questa technology provides a comprehensive verification solution for high-speed interconnects, and includes models, coverage, checkers, sequences, and test plans.
  • The Tessent™ software DFT platform, which helped Graphcore address a broad spectrum of DFT challenges associated with producing a device of the size and complexity of the Colossus GC200 IPU. Using Tessent SiliconInsight™ software for benchtop testing, Graphcore validated all onboard logic BIST, ATPG and memory BIST elements well ahead of schedule. 
  • To develop the M2000 platform around the Colossus processors, Graphcore is leveraging Mentor solutions to address important power and thermal management challenges posed by such a dense design, enabling them to optimize performance and dramatically reduce design cycles. Graphcore designed the PCBs using Mentor’s Xpedition™ software concurrent team design platform, and verified the system for performance and manufacturability during the PCB design process with Mentor’s HyperLynx™ software DRC, Valor™ software NPI, and Simcenter™ Flotherm™ software.

“Demand for increasingly sophisticated AI chips is driving the industry’s top EDA software providers to enhance the capacity and scalability of their offerings across the entire design flow,” said Adrian Buckley, vice president, Europe for Mentor. “Customers like Graphcore are turning to Mentor’s highly advanced tools to help bring their products to market. Mentor is pleased to have played such a significant role in development of Graphcore’s ColossusGC200 IPU, which is one of the largest and most advanced AI platforms ever developed.”

Leave a Reply

featured blogs
Nov 24, 2020
The ICADVM20.1 and IC6.1.8 ISR15 production releases are now available for download at Cadence Downloads . For information on supported platforms and other release compatibility information, see the... [[ Click on the title to access the full blog on the Cadence Community si...
Nov 23, 2020
It'€™s been a long time since I performed Karnaugh map minimizations by hand. As a result, on my first pass, I missed a couple of obvious optimizations....
Nov 23, 2020
Readers of the Samtec blog know we are always talking about next-gen speed. Current channels rates are running at 56 Gbps PAM4. However, system designers are starting to look at 112 Gbps PAM4 data rates. Intuition would say that bleeding edge data rates like 112 Gbps PAM4 onl...
Nov 20, 2020
[From the last episode: We looked at neuromorphic machine learning, which is intended to act more like the brain does.] Our last topic to cover on learning (ML) is about training. We talked about supervised learning, which means we'€™re training a model based on a bunch of ...

featured video

AI SoC Chats: Protecting Data with Security IP

Sponsored by Synopsys

Understand the threat profiles and security trends for AI SoC applications, including how laws and regulations are changing to protect the private information and data of users. Secure boot, secure debug, and secure communication for neural network engines is critical. Learn how DesignWare Security IP and Hardware Root of Trust can help designers create a secure enclave on the SoC and update software remotely.

Click here for more information about Security IP

featured paper

Overcoming PPA and Productivity Challenges of New Age ICs with Mixed Placement Innovation

Sponsored by Cadence Design Systems

With the increase in the number of on-chip storage elements, it has become extremely time consuming to come up with an optimized floorplan using manual methods, directly impacting tapeout schedules and power, performance, and area (PPA). In this white paper, learn how a breakthrough technology addresses design productivity along with design quality improvements for macro-dominated designs. Download white paper.

Click here to download the whitepaper

Featured Chalk Talk

Microchip SAM11L KPH

Sponsored by Mouser Electronics and Microchip

Adding connectivity to your embedded design opens up a whole new realm of security challenges. Inviting your device to the IoT requires careful attention to building a secure foundation. In this episode of Chalk Talk, Amelia Dalton chats with Anand Rangarajan from Microchip about the SAML11-KPH MCU and how it can help you develop your application without worrying about security.

Click here for more information about Microchip Technology SAM L10/L11 ARM® Cortex®-M23 MCUs