A look at the power benefits and performance impact as designs move closer to voltage thresholds.
Lauri Koskinen, CTO and founder of Minima Processor, and Ron Moore, vice president of marketing at ARM, talk about near-threshold computing, dynamic power and margining, and how these techniques can extend battery life and reduce energy consumption.
The industry is gaining ground in understanding how aging affects reliability, but more variables make it harder to fix.
Tools become more specific for Si/SiGe stacks, 3D NAND, and bonded wafer pairs.
Thinner photoresist layers, line roughness, and stochastic defects add new problems for the angstrom generation of chips.
Key pivot and innovation points in semiconductor manufacturing.
The verification of a processor is a lot more complex than a comparably-sized ASIC, and RISC-V processors take this to another layer of complexity.
Less precision equals lower power, but standards are required to make this work.
New applications require a deep understanding of the tradeoffs for different types of DRAM.
127 startups raise $2.6B; data center connectivity, quantum computing, and batteries draw big funding.
The industry is gaining ground in understanding how aging affects reliability, but more variables make it harder to fix.
Thermal mismatch in heterogeneous designs, different use cases, can impact everything from accelerated aging to warpage and system failures.
Technical and business challenges persist, but momentum is building.
The verification of a processor is a lot more complex than a comparably-sized ASIC, and RISC-V processors take this to another layer of complexity.
New memory standard adds significant benefits, but it’s still expensive and complicated to use. That could change.
And then your business guy says, we will never be able to sell this 🙂 Well explained!
I went to a tech paper on that at DAC in SF a few years ago (IBM I think), so nothing new.
Unfortunately nobody is fixing the tool flow for doing verification of variable/low Voltage stuff (DVFS etc.), and you really want to be doing asynchronous circuits too which VCS/NC etc. definitely don’t handle.
These two things can fix that –
https://xyce.sandia.gov/
http://www.wiley.com/WileyCDA/WileyTitle/productCd-0470226099,subjectCd-EEJ0.html
It’s just a question of who wants to put some money into the tools, and I haven’t seen ARM do anything on that front, and I wouldn’t hold my breath waiting for any of the EDA companies to fix things (given past performance on AMS issues).
However, the prize here is quite large – not so much the extended battery life, but the ability to die-stack many layers of Silicon to create very dense computing machines that start getting close to what the human brain can do. NB: good if you can do thermal modeling too for that.