RSA-ECC Public Key Accelerator Engine, 2K ops/sec, DPA Resistant
AI expands HBM footprint
By Gary Hilson, EETimes (January 20, 2022)
High bandwidth memory (HBM) is becoming more mainstream. With the latest iteration’s specifications approved, vendors in the ecosystem are gearing to make sure it can be implemented so customers can begin to design, test and deploy systems.
The massive growth and diversity in artificial intelligence (AI) means HBM is less than niche. It’s even become less expensive, but it’s still a premium memory and requires expertise to implement. As a memory interface for 3D-stacked DRAM, HBM achieves higher bandwidth while using less power in a form factor that’s significantly smaller than DDR4 or GDDR5 by stacking as many as eight DRAM dies with an optional base die which can include buffer circuitry and test logic.
Like all memory, HBM makes advances in performance improvement and power consumption with every iteration. A key change when moving to HBM3 from HBM2 will a 100% performance improvement in the data transfer rate from 3.2/3.6Gbps to 6.4Gbps max per pin, said Jinhyun Kim, principal engineer with Samsung Electronics’ memory product planning team.
A second fundamental change is a 50% increase in the maximum capacity from 16GB (8H) to 24GB (12H). Finally, HBM3 implements on-die error correction code as an industry-wide standard, which improves system reliability, Kim said. “This will be critical for the next generation of artificial intelligence and machine learning systems.”
E-mail This Article | Printer-Friendly Page |
Related News
- MIPS Expands Global Footprint with New Design Center and Talent for Systems Architects and AI Compute
- Ceva Expands AI Ecosystem for its Class-Leading NPU IP with New Partnerships for Automotive and Edge AI
- Cadence Expands Tensilica IP Portfolio with New HiFi and Vision DSPs for Pervasive Intelligence and Edge AI Inference
- Thalia expands footprint in China with appointment of Business Partner Fionn Liu
- Alphawave Semi Expands Collaboration with Samsung, Adds 3nm Connectivity IP to Meet Accelerated AI and Data Center Demand
Breaking News
- lowRISC Deploys Real Intent Ascent Lint, Meridian CDC, & Meridian RDC for OpenTitan Project
- Silicon Proven PCIe 5.0 PHY and Controller IP Cores in 12nm to Revolutionize Connectivity solutions
- Arm revenues up 47%; shares fall
- Sondrel awarded new Video Processor ASIC design and supply contract for a leading provider of High-Performance Video systems
- X-Silicon Announces a NEW Low-Power Open-Standard Vulkan-Enabled C-GPU™ - a RISC-V Vector CPU Infused with GPU ISA and AI/ML acceleration in a Single Processor Core
Most Popular
- Arm revenues up 47%; shares fall
- X-Silicon Announces a NEW Low-Power Open-Standard Vulkan-Enabled C-GPU™ - a RISC-V Vector CPU Infused with GPU ISA and AI/ML acceleration in a Single Processor Core
- Softbank reported to be in talks to buy Graphcore
- Siemens delivers end-to-end silicon quality assurance for next-generation IC designs with new Solido IP Validation Suite
- Ceva, Inc. Announces First Quarter 2024 Financial Results