HiPEAC Conference 2022
Budapest, 20th – 22nd June 2022
The HiPEAC conference is the premier European forum for experts in computer architecture, programming models, compilers and operating systems for general-purpose, embedded and cyber-physical systems. Areas of focus and integration include safety-critical dependencies, cybersecurity, energy efficiency and machine learning.
The HiPEAC 2022 conference will take place in Budapest, Hungary. Associated workshops, tutorials, special sessions, several large poster session and an industrial exhibition will run in parallel with the conference.
Keynotes
Hai Li
Duke University, United States Mon, June 20th: 9:00 – 10:00
Efficient Machine Learning: Algorithms-Hardware Co-design
Following technology advances in high performance computation systems and fast growth of data acquisition, machine learning, especially deep neural networks (DNNs), made remarkable success in many research areas and applications. Such a success, to a great extent, is enabled by developing large-scale network models that learn from a huge volume of data. The deployment of such a big model, however, is both computation-intensive and memory-intensive. Though the research on hardware acceleration for neural network has been extensively studied, the progress of hardware development still falls far behind the upscaling of DNN models at software level. The holistic co-design across algorithm, architecture, and device levels emerges more important for execution acceleration, energy efficiency, and design flexibility. In this presentation, we will present our studies on how to optimize the training process for sparse and low-precision network models for general platforms. We will also discuss the neural architectural search optimized for DNN operations.
Ovidiu Vermesan
SINTEF, Norway Tue, June 21st: 9:00 – 10:00
Edge Distributed Intelligence in the Internet of Autonomous Systems
The convergence of artificial intelligence (AI) enabled edge processing, and the Internet of Things (IoT) enhances the cognitive capabilities of distributed autonomous systems offering new paradigms for addressing the collective behavior of self-organized autonomous systems. In this context, integrating AI-based methods such as machine learning (ML) and deep learning (DL) requires new concepts for energy-efficient and scalable solutions across the edge continuum. The presentation gives an overview of the future technology trends in the field, emphasizing the technology development roadmap for the following years.
Bianca Schroeder
University of Toronto, Canada Wed, June 22nd: 9:00 – 10:00
Operational Characteristics of Flash-Based Storage in Production Systems
Over the past decade flash-based solid state drives have became indispensible in modern storage systems. In this talk we will look at several years worth of field data to understand how these devices behave in production deployments, how they compare to HDDs and explore various aspects where reality might differ from expectations. In particular, we focus on reliability and how it is affected by factors such as drive age, firmware, flash technology, and also other operational aspects, such as the level of write amplification experienced by SSDs in production systems and how it is affected by various factors; the effectiveness of wear leveling; or the rate at which drives in the field use up their program- erase (PE) cycle limit and what that means for the transition to future generations of flash with lower endurance.