eSilicon to debut AI Accelerator software and a new chiplet model at Hot Chips 2019
New software maps high-level AI workloads to eSilicon’s neuASIC modular AI ASIC IP platform
August 15, 2019 -- San Jose, Calif. -- eSilicon, an independent provider of FinFET-class ASIC design, market-specific IP platforms and advanced 2.5D packaging solutions, will exhibit at Hot Chips 31.
eSilicon at Hot Chips 31
Introducing AI Accelerator Software
eSilicon’s 7nm FinFET neuASIC™ AI IP platform is a library of AI-specific tiles that can be configured to support an AI algorithm. eSilicon’s new AI Accelerator maps high-level AI workloads to the neuASIC platform and estimates PPA (power, performance, area) for the algorithm in the resultant silicon implementation.
neuASIC IP plus the AI Accelerator software allow design exploration of candidate architectures to ensure the design will be within the target specifications. This approach supports changes to the algorithm or the package.
eSilicon will be demonstrating AI Accelerator on Monday and Tuesday at its sponsor table.
AI Accelerator is available in IP Navigator, eSilicon’s IP exploration and evaluation tool, at no charge. A free eSilicon STAR account is required to access Navigator, which may be requested here.
Update on Chiplets
The idea of chiplets makes sense in terms of yield, cost and risk, but they haven’t really caught on. Is there a chiplet business model that works?
When & Where
Sunday-Tuesday, August 18-20, 2019
Memorial Auditorium, Stanford Campus
Palo Alto, California
Registration for the conference is still open online or on site during the event at the registration booth.
About eSilicon
eSilicon provides complex FinFET ASICs, market-specific IP platforms and advanced 2.5D packaging solutions. Our ASIC-proven, differentiating IP includes highly configurable 7nm 56G/112G SerDes plus networking-optimized 16/14/7nm FinFET IP platforms featuring HBM2 PHY, TCAM, specialized memory compilers and I/O libraries. Our neuASIC platform provides AI-specific IP and a modular design methodology to create adaptable, highly efficient AI ASICs. eSilicon serves the high-bandwidth networking, high-performance computing, AI and 5G infrastructure markets. www.esilicon.com
Related Semiconductor IP
- HBM4 PHY IP
- Ultra-Low-Power LPDDR3/LPDDR2/DDR3L Combo Subsystem
- MIPI D-PHY and FPD-Link (LVDS) Combinational Transmitter for TSMC 22nm ULP
- HBM4 Controller IP
- IPSEC AES-256-GCM (Standalone IPsec)
Related News
- Cassia.ai Achieves Breakthrough in AI Accelerator Technology with Successful Tapeout of two Test Chips
- AI Hardware Summit 2019: Booth-to-Booth eSilicon 58G DSP-Based SerDes Demonstration Over a Five-Meter Samtec Copper Cable
- ECOC 2019: eSilicon to Demonstrate 7nm 58G DSP-Based SerDes Over Seven-Meter and Three-Meter Samtec Cable Assemblies
- NEO Semiconductor Introduces World’s First Extreme High Bandwidth Memory (X-HBM) Architecture for AI Chips
Latest News
- Weebit Nano secures a license agreement with Texas Instruments
- Digital Blocks DB9000 Display Controller IP Core Family Extends Leadership in 8K, Automotive, Medical, Aerospace, and Industrial SoC Designs
- AI Directs UFS Advancement
- Qualitas Semiconductor Expands Automotive Momentum with 5nm IP Bundle Agreement
- Cyient Semiconductors Acquires Majority Stake in Kinetic Technologies to Drive Custom Power IC Leadership for Edge AI and High-Performance Compute Markets