Researchers have studied neural network compression for quite some time. However, the need for always on compute has led to a recent trend towards executing these applications on even smaller IoT devices…
Arm-University of Bristol Centre of Excellence
Following many years of collaboration on a range of research projects, Arm Research and the University of Bristol have established an Arm Centre of Excellence. Six months in, the Centre of Excellence is…
Arm Research Summit: Five Years of Innovation and Collaboration
The Arm Research Summit has been engaging academics, industry experts and researchers since 2016, promoting collaboration in a research led environment. Overtime the Summit has grown and developed, seeing…
A Year of Discovery: Arm Research 2019
2019 was yet another year of incredible technology discovery for Arm Research. Inspiring advancements have been made across the research community, and Arm Research has contributed to this. Our teams have…
Research Enablement: Helping you achieve your research goals
The Arm Research Enablement team announced some exciting new initiatives at the Arm Research Summit 2020, focusing on the needs of the academic research community. Enablement are dedicated to working closely…
A Low-power Timer Enabling Timed Shutdown of Arm-based Microcontrollers for IoT Applications
IoT sensor nodes place severe energy constraints on computing platforms. Duty-cycling helps to reduce the power of the sensor node to save energy, but requires an ultra-low-power timer for reliable timed…
Arm Research Summit 2020 | Global Technology Challenges
The Arm Research Summit fuels collaboration, research and innovation with delegates from a range of research, academic and industrial backgrounds. The Arm Research Summit 2020 will take place online, 9…
Sparse Architecture Search (SpArSe): Democratizing and Enabling TinyML on Arm M-class
Microcontrollers (MCUs) are the ubiquitous computer of our time, being tiny, cheap, and low power. Often powered by using solar cell, they are in your watch, fridge, and your car will contain about thirty…
Skipping RNN State-updates Without Retraining the Original Model
Recurrent Neural Networks (RNNs) are used in tasks where the strict order of the input conveys certain information, and so are classed as an important algorithm. These networks are being deployed on resource…
Collaboration Case Study: Machine Learning Hardware with Harvard University
University engagements play a significant role in our partnerships at Arm Research, helping us extend our reach and build understanding of technologies as they emerge. Harvard University is one of those…
Wireless 3D Integration – Making Stacking Silicon as Easy as Stacking Lego (2/2)
The previous instalment of this blog discussed the benefits that contactless - or wireless - 3D integration can offer in an Internet of Things (IoT) context. But how are they designed? This blog dives…
Wireless 3D Integration – Making Stacking Silicon as Easy as Stacking Lego (1/2)
The growth of the Internet of Things has brought about a range of new challenges for IC designers, now IoT devices are required to perform a range of different functions, within strict power and volume…
Emerging Standalone NVM – Moving Beyond the Hype
Non-volatile memory (NVM) has been discussed in numerous articles and papers over the last few decades, and hyped as either a DRAM replacement in the optimistic case, a Flash replacement, or a new memory…
Arm Research Summit 2019: Catch Up and Highlights
The fourth annual Arm Research Summit, and the first in the US, held in Austin, Texas, created a collaborative space for academics, researchers and industry experts globally to discuss complex computing…
The ICs of 2030
This blog is a companion to my keynote at Arm TechCon on 10 October, 2019. I am publishing a blog form because I cover a LOT of ground in that 20 minute talk, and there are i...
Continuous Cross-Architecture Integration with GitLab
Arm has had a long history of research into distributed systems. We have previously focused that research on applying Arm technology to HPC, but in recent years we’ve also been looking at various forms…
Arm Flexible Access for Research to Accelerate Innovation for Academia and Research
In July of this year, Arm announced the availability of Arm Flexible Access which expands how partners can access and license Arm technology for semiconductor design. This new engagement model is already…
Tasking Lives Up To Its Promises
Tasking could help overcome a major resource underutilization issue in High-Performance Computing, When parallel applications partition and distribute compute and data across processors in the system,…
Characterization of Multi-threaded HPC Codes
Core counts continue to increase for High-Performance Computing (HPC) systems, but multiple factors may prevent current software from fully utilizing the increased available thread count. Inter-thread…
Creating Minimal Docker Images for the Edge
Carefully crafting your Dockerfiles proves to be not only important for minimizing your container attack surface, but also for speeding up the deployment times of applications running both in the cloud…
Alpha-Blending: Quantizing networks without using the STE
Increasingly, intelligent applications are using neural networks at their core to deliver new functionality to users. These applications include language understanding and translation, image recognition…