Hyderabad, Telangana, India:
Ceremorphic® Inc., developer of probably the most dependable and vitality AI supercomputing structure, introduced immediately that it has efficiently taped out its first 5nm chip with its companion TSMC. As the one AI supercomputing chip to unravel the high-performance computing wants in reliability, safety and energy consumption at scale, the Ceremorphic answer delivers the efficiency required for next-generation functions resembling AI mannequin coaching, HPC, drug discovery and metaverse processing.
“Ceremorphic is the first company to tape out an AI supercomputing chip in the advanced 5nm process node, representing not only a significant accomplishment for our company, but also for the India semiconductor industry,” stated Dr. Venkat Mattela, Founder and CEO of Ceremorphic. “The talent pool, expertise and resulting innovation coming from India has once again proven to be world-class and this design achievement positions us well for our next phase of growth and expansion.”
“With this announcement, the AI industry just got one step closer to finally addressing today’s critical challenges through reliable performance computing,” stated Subhasish Mitra, Professor of Electrical Engineering and of Computer Science at Stanford University. “Congratulations to the entire Ceremorphic team for achieving this significant milestone and bringing a new level of innovation to the AI supercomputing community.”
About the Ceremorphic Architecture
The Ceremorphic chip incorporates dependable analog and digital circuits, very excessive velocity (64Gbit, PCIe 6.0) and vitality environment friendly connectivity circuits for system degree and inter-node communications, multi-thread dependable processor and low energy circuits for SoC silicon. This Hierarchical Learning Processor (HLP) deploys the suitable processing system for optimum energy efficiency operation. Key options of the chip embody the next:
- Patented Multi-thread processing macro-architecture, ThreadArch® primarily based RISC –V® processor for proxy processing (1GHz)
- Custom designed X16 PCIe 6. 0 / CXL 3.0 connectivity interface
- Reliability system parts realized in superior microarchitectures
Ceremorphic in India
Ceremorphic’s improvement middle in India at the moment employs greater than 150 {hardware} and software program engineers with plans to develop to 250 by 2023. Earlier this yr, the corporate introduced a $50 million Series A spherical and plans to innovate the AI supercomputing house with an vitality environment friendly, excessive efficiency chip structure.
Founded in April 2020, Ceremorphic at the moment has 150 full-time workers devoted to creating superior silicon and software program merchandise for next-generation computing techniques. Leveraging greater than 120 patents and confirmed experience in creating industry-leading silicon system merchandise, Ceremorphic has constructed a brand new structure that delivers the efficiency wanted for functions resembling AI mannequin coaching, automotive processing, drug discovery, HPC, and metaverse processing with unprecedented vitality effectivity and reliability. Designed in superior silicon know-how node (TSMC 5nm), this structure solves immediately’s high-performance computing issues in reliability, safety and vitality consumption throughout all performance-demanding market segments. For extra info, go to https://www.ceremorphic.com.
Copyright 2022. All rights reserved. Ceremorphic and ThreadArch are registered emblems and Reliable Performance Computing and Hierarchical Learning Processor are emblems of Ceremorphic, Inc. All different emblems are the property of their respective homeowners.
Click right here for Media Contact Details
Submit your press launch
