Memory Systems for AI: Part 1
There has been quite a lot of recent news about domain-specific processors that are being designed for the artificial intelligence (AI) market. Interestingly, many of the techniques used today in modern AI chips and applications have actually been around for several decades. However, neural networks didn’t really take off during the last wave of interest in AI that spanned the 1980s and 1990s. The question is why.
The chart above provides some insight as to why AI technology remained relatively static for so many years. Back in the 1980s and 1990s, processors (CPUs) simply weren’t fast enough to adequately handle AI applications. In addition, memory performance wasn’t yet good enough to enable neural networks and modern techniques to displace conventional approaches. Consequently, conventional approaches remained popular in the above-mentioned timeframe.
To read the full article, click here
Related Semiconductor IP
- 1G to 100G Single-Port MACsec Engine
- 1.6T/3.2T Multi-Channel MACsec Engine with TDM Interface (MACsec-IP-364)
- Fast NIST ESV certified, FIPS (SP800-90A/B/C) True Random Number Generator
- Programmable Root of Trust Family With DPA & Quantum Safe Cryptography
- AES XTS/GCM Accelerators
Related Blogs
- Enabling Memory Choice for Modern AI Systems: Tenstorrent and Rambus Deliver Flexible, Power-Efficient Solutions
- LPDDR6: A New Standard and Memory Choice for AI Data Center Applications
- The Memory Imperative for Next-Generation AI Accelerator SoCs
- The Silent Guardian of AI Compute - PUFrt Unifies Hardware Security and Memory Repair to Build the Trust Foundation for AI Factories
Latest Blogs
- IDS-NoC: A Scalable Interconnect Solution for Modern SoC Design
- Leading-edge AI IC designs demand comprehensive HAV methodologies
- Beyond the Fab: Building Europe’s Next Generation of Semiconductor Champions
- AI Semiconductor Design at 3nm and 2nm: Silicon-Proven IP
- Samsung Foundry and Cadence Expand Infrastructure and Physical AI Solution