via Stanford University
Today’s electronics waste energy by using separate chips to process and store data. New algorithms enable energy-efficient hybrids to work together as one mega-chip.
Smartwatches and other battery-powered electronics would be even smarter if they could run AI algorithms.
But efforts to build AI-capable chips for mobile devices have so far hit a wall — the so-called “memory wall” that separates data processing and memory chips that must work together to meet the massive and continually growing computational demands imposed by AI.
“Transactions between processors and memory can consume 95 percent of the energy needed to do machine learning and AI, and that severely limits battery life,” said computer scientist Subhasish Mitra, senior author of a new study published in Nature Electronics.
Now, a team that includes Stanford School of Engineering computer scientist Mary Wootters and electrical engineer H.-S. Philip Wong has designed a system that can run AI tasks faster, and with less energy, by harnessing eight hybrid chips, each with its own data processor built right next to its own memory storage.
This paper builds on the team’s prior development of a new memory technology, called RRAM, that stores data even when power is switched off — like flash memory — only faster and more energy efficiently. Their RRAM advance enabled the Stanford researchers to develop an earlier generation of hybrid chips that worked alone. Their latest design incorporates a critical new element: algorithms that meld the eight, separate hybrid chips into one energy-efficient AI-processing engine.
“If we could have built one massive, conventional chip with all the processing and memory needed, we’d have done so, but the amount of data it takes to solve AI problems makes that a dream,” Mitra said. “Instead, we trick the hybrids into thinking they’re one chip, which is why we call this the Illusion System.”
The researchers developed Illusion as part of the Electronics Resurgence Initiative (ERI), a $1.5 billion program sponsored by the Defense Advanced Research Projects Agency. DARPA, which helped spawn the internet more than 50 years ago, is supporting research investigating workarounds to Moore’s Law, which has driven electronic advances by shrinking transistors. But transistors can’t keep shrinking forever.
“To surpass the limits of conventional electronics, we’ll need new hardware technologies and new ideas about how to use them,” Wootters said.
The Stanford-led team built and tested its prototype with help from collaborators at the French research institute CEA-Leti and at Nanyang Technological University in Singapore. The team’s eight-chip system is just the beginning. In simulations, the researchers showed how systems with 64 hybrid chips could run AI applications seven times faster than current processors, using one-seventh as much energy.
Such capabilities could one day enable Illusion Systems to become the brains of augmented and virtual reality glasses that would use deep neural networks to learn by spotting objects and people in the environment, and provide wearers with contextual information — imagine an AR/VR system to help birdwatchers identify unknown specimens.
Stanford graduate student Robert Radway, who is first author of the Nature Electronics study, said the team also developed new algorithms to recompile existing AI programs, written for today’s processors, to run on the new multi-chip systems. Collaborators from Facebook helped the team test AI programs that validated their efforts. Next steps include increasing the processing and memory capabilities of individual hybrid chips and demonstrating how to mass produce them cheaply.
“The fact that our fabricated prototype is working as we expected suggests we’re on the right track,” said Wong, who believes Illusion Systems could be ready for marketability within three to five years.
The Latest Updates from Bing News & Google News
Go deeper with Bing News on:
- Intel to spend $3.5B on New Mexico fabs and $600M on chip R&D centers in Israelon May 3, 2021 at 8:23 pm
Intel Corp. confirmed today it’s planning to invest $600 million to expand its research and development facilities in Israel, in addition to the $10 billion it has already committed toward building a ...
- Intel To Invest $600M In ‘Mega Chip’ Campus, Mobileye R&D Center In Israelon May 3, 2021 at 6:30 am
Intel will invest $200m in its facility in Haifa and $400m for Mobileye's new center; CEO hails a 'vibrant future for Intel, Israel for decades to come.' ...
- Intel to invest $600 million in massive Israel expansionon May 3, 2021 at 2:07 am
Following the global computing giant's announcement last week that it would build a $200 million campus in Haifa, Intel Global plans to invest millions more as they expand their Israeli presence.
- Intel to invest $600 mln to expand chip, Mobileye R&D in Israelon May 2, 2021 at 9:40 pm
JERUSALEM: Intel Corp said on Sunday it will invest another $600 million in Israel to expand its research and development (R&D) and confirmed it was spending $10 billion on a new chip plant. The ...
- Intel to invest billions in chip fabrication, R&D in Israelon May 2, 2021 at 6:01 pm
Intel will invest another US$600 million (A$777 million) in Israel to expand its research and development (R&D) and confirmed it was spending US$10 billion on a new chip plant. The announcement was ...
Go deeper with Google Headlines on:
Go deeper with Bing News on:
- A new 1P1R Image Sensor with In-Memory Computing Properties based on Silicon Nitride Deviceson May 5, 2021 at 12:31 am
Research progress in edge computing hardware, capable of demanding in-the-field processing tasks with simultaneous memory and low power properties, is leading the way towards a revolution in IoT ...
- RRAM Market with a Comprehend Key Patterns & Development in Impact of Covid-19 with CAGR of +10% by 2027:PSCS, Adesto, Crossbar, Fujitsu, Inon May 3, 2021 at 12:45 am
This report on the global RRAM market assures immense details on a range of growth opportunities across the forecast period of 2021-2027. The report has extensive details by skilled experts. The study ...
- CXL: Coherency, Memory, and I/O Semantics on PCIe Infrastructureon April 28, 2021 at 1:37 pm
Compute Express Link is a cache-coherent interconnect for processors, memory expansion, and accelerators that maintains a unified coherent memory space between the CPU and any memory on the attached ...
- Variability and Energy Consumption Tradeoffs in Multilevel Programming of RRAM Arrayson April 26, 2021 at 10:15 pm
Achieving a reliable multilevel programming operation in resistive random access memory (RRAM) arrays is still a challenging task. In this work, we assessed the impact of the voltage step value used ...
- Transparent Electronics Market Growth, Share, Demand and Applications Forecast to 2025on April 22, 2021 at 2:31 pm
The development of Resistive Random-Access Memory (RRAM) based on zinc tin oxide is also noted as a memristor which helps the computer memory operate on resistance and is expected to open a new avenue ...