草榴社区

Accelerating Development of a Novel Analog In-Memory Compute AI Accelerator SoC Using 草榴社区 Cloud

TetraMem Inc, a US-based company founded in Silicon Valley in 2018, is poised to deliver the industry’s most disruptive inmemory computing (IMC) technology for efficient AI applications. TetraMem is the world’s only company to produce a high bit-density multi-level RRAM (aka computing memristor) based accelerator in commercial foundries. The technology was featured in the March 30, 2023 edition of Nature. Our team brings together complementary skill sets and technological know-how ranging from proprietary computing memory devices to novel computing architectures.

CHALLENGES

  • Having a global team spread across continents requires a seamless, unified mixed-signal SoC design environment for efficient collaboration
  • CAD setup/flows and on-demand scalable and elastic IT resources are needed
  • The AI accelerator’s complex design requires smooth integration and verification of analog in-memory computing and the digital RISC-V processor, with critical time-to-market targets

 

SYNOPSYS SOLUTION

  • Pre-configured end-to-end analog design flow helped jumpstart development
  • Intuitive and user-friendly browser-based platform for chip design
  • On-demand pay-per-use access to unlimited EDA resources without licensing constraints, enabling faster time to results
  • Instant availability of scalable and cutting-edge compute and storage

 

BENEFITS 

  • Production environment: Get up and running in days instead of weeks
  • Complete and intuitive design flow: Providing simulation and verification capabilities for designs, eliminating the need for CAD setup, and enabling automated PDK and technology setup
  • Improved collaboration and productivity: Global teams can collaborate on the same schematic, layout, or design step in real time
  • On-demand, pay-per-use access: Scalable compute and unlimited tool licenses speed time to market, enabling achievement of PPA targets and robust design by allowing more iterations in less time
  • Availability: Includes the entire portfolio of 草榴社区 EDA software and high-quality, silicon-proven 草榴社区 IP

We were able to achieve a very fast infrastructure setup on the 草榴社区 Cloud EDA environment within days."

Wenbo Yin

|

VP of IC Design, TetraMem Inc.

ABOUT SYNOPSYS CLOUD

草榴社区 Cloud combines the availability of advanced computing and storage infrastructure with unlimited access to EDA software licenses on demand so you can focus on what you do best—design chips—faster. With cloud-native EDA tools and pre-optimized hardware platforms, an extremely flexible business model, and a modern customer experience, 草榴社区 has reimagined the future of chip design on the cloud that doesn’t disrupt proven workflows.

The 草榴社区 Cloud FlexEDA business model offers two licensing options: pay-per-use (PPU) and cloud subscription license (CSL). PPU is an industry-first, true usage-based licensing approach for EDA tools. 草榴社区 Cloud FlexEDA provides access to unlimited, on-demand EDA software licenses, which is a transformational change compared to traditional EDA software licensing models. With 草榴社区 Cloud FlexEDA, many 草榴社区 tools are now available for use by the minute, providing customers with the granularity they need for peak usage bursts in the cloud. This helps reduce time to results significantly and deliver a better quality design ahead of time.

Users can choose from two deployment options: Bring-Your-Own-Cloud (BYOC) and Software-as-a-Service (SaaS). 草榴社区 Cloud offers the flexibility to use either one or both deployment options, depending on customer requirements. 

Request a Deep-dive Technical Session with a Subject Matter Expert