Section: Partnerships and Cooperations
Participants : Olivier Sentieys, Silviu-Ioan Filip.
The design and implementation of convolutional neural networks for deep learning is currently receiving a lot of attention from both industrials and academics. However, the computational workload involved with CNNs is often out of reach for low power embedded devices and is still very costly when run on datacenters. By relaxing the need for fully precise operations, approximate computing substantially improves performance and energy efficiency. Deep learning is very relevant in this context, since playing with the accuracy to reach adequate computations will significantly enhance performance, while keeping quality of results in a user-constrained range. AdequateDL will explore how approximations can improve performance and energy efficiency of hardware accelerators in deep-learning applications. Outcomes include a framework for accuracy exploration and the demonstration of order-of-magnitude gains in performance and energy efficiency of the proposed adequate accelerators with regards to conventional CPU/GPU computing platforms.
Participants : Olivier Sentieys, Cédric Killian, Joel Ortiz Sosa.
The efficient exploitation by software developers of multi/many-core architectures is tricky, especially when the specificities of the machine are visible to the application software. To limit the dependencies to the architecture, the generally accepted vision of the parallelism assumes a coherent shared memory and a few, either point to point or collective, synchronization primitives. However, because of the difference of speed between the processors and the main memory, fast and small dedicated hardware controlled memories containing copies of parts of the main memory (a.k.a caches) are used. Keeping these distributed copies up-to-date and synchronize the accesses to shared data, requires to distribute and share information between some may if not all the nodes. By nature, radio communications provide broadcast capabilities at negligible latency, they have thus the potential to disseminate information very quickly at the scale of a circuit and thus to be an opening for solving these issues. In the RAKES project, we intend to study how wireless communications can solve the scalability of the abovementioned problems, by using mixed wired/wireless Network on Chip. We plan to study several alternatives and to provide (a) a virtual platform for evaluation of the solutions and (b) an actual implementation of the solutions.
Participants : Olivier Sentieys, Cédric Killian, Daniel Chillet.
The aim of Opticall is to design broadcast-enabled optical communication links in manycore architectures at wavelengths around 1.3m. We aim to fabricate an optical broadcast link for which the optical power is equally shared by all the destinations using design techniques (different diode absorption lengths, trade-off depending on the current point in the circuit and the insertion losses). No optical switches will be used, which will allow the link latency to be minimized and will lead to deterministic communication times, which are both key features for efficient cache coherence protocols. The second main objective of Opticall is to propose and design a new broadcast-aware cache coherence communication protocol allowing hundreds of computing clusters and memories to be interconnected, which is well adapted to the broadcast-enabled optical communication links. We expect better performance for the parallel execution of benchmark programs, and lower overall power consumption, specifically that due to invalidation or update messages.
Participants : Cédric Killian, Daniel Chillet, Olivier Sentieys, Emmanuel Casseau.
The goal of the SHNoC project is to tackle one of the manycore interconnect issues (scalability in terms of energy consumption and latency provided by the communication medium) by mixing emerging technologies. Technology evolution has allowed for the integration of silicon photonics and wireless on-chip communications, creating Optical and Wireless NoCs (ONoCs and WNoCs, respectively) paradigms. The recent publications highlight advantages and drawbacks for each technology: WNoCs are efficient for broadcast, ONoCs have low latency and high integrated density (throughput/cm) but are inefficient in multicast, while ENoCs are still the most efficient solution for small/average NoC size. The first contribution of this project is to study the compatibility of processes to associate the three aforementioned technologies and to define an hybrid topology of the interconnection architecture. This exploration will determine the number of antennas for the WNoC, the amount of embedded lasers sources for the ONoC and the routers architecture for the ENoC. The second main contribution is to provide quality of service of communication by determining, at run-time, the best path among the three NoCs with respect to a target, e.g. minimizing the latency or energy. We expect to demonstrate that the three technologies are more efficient when jointly used and combined, with respect to traffic characteristics between cores and quality of service targeted.
Participants : Davide Pala, Olivier Sentieys.
The ZEP project addresses the issue of designing tiny, batteryless, computing objects harvesting energy in the environment. The main application target is Internet of Things (IoT) where small communicating objects will be composed of this computing part associated to a low-power wake-up radio system. The energy level harvested being very low, very frequent energy shortages are expected, which makes the systems following the paradigm of Intermittently-Powered Systems. In order for the system to maintain a consistent state, it will be based on a new architecture embedding non-volatile memory (NVRAM). The major outcomes of the project will be a prototype harvesting board including NVRAM and the design of a new non-volatile processor (NVP) associated with its optimizing compiler and operating system. Cairn is focusing on the microarchitecture of the NVP and on new strategies for backup and restore data and processor state. The ZEP project gathers four Inria teams that have a scientific background in architecture, compilation, operating system and low power together with the CEA Grenoble. Another important goal of the project is to structure the research and innovation that should occur within Inria to prepare the important technological shift brought by NVRAM technologies.
DGA RAPID - FLODAM (2017–2021)
Participants : Joseph Paturel, Simon Rokicki, Olivier Sentieys, Angeliki Kritikakou.
FLODAM is an industrial research project for methodologies and tools dedicated to the hardening of embedded multi-core processor architectures. The goal is to: 1) evaluate the impact of the natural or artificial environments on the resistance of the system components to faults based on models that reflect the reality of the system environment, 2) the exploration of architecture solutions to make the multi-core architectures fault tolerant to transient or permanent faults, and 3) test and evaluate the proposed fault tolerant architecture solutions and compare the results under different scenarios provided by the fault models. For more details see https://flodam.fr