Qarnot computing aims to develop a flexible High Performance Computing grid able to reuse the heat emitted by calculation units for the heating purposes of houses and business premises. Regularly, huge data centers are opened for storage or calculation of numerical data. Needs, including calculation power, are becoming increasingly important, which explains the creation and implementation of more and more of these data centers. However, the past few years have shown that these clusters are particularly energy-consuming not only because of the calculations but also because of the cooling systems. The idea is to sale processing power on the cloud. The calculations are smartly distributed to a remote computing grid. The grid is composed of calculation units that are actually electric heaters. The heat is produced by the processors that perform the computing. Qarnot computing sales cloud HPC on one end, and refunds the electricity used by electric heaters on the other end. This solution allows Qarnot to build progressively a scalable computing grid avoiding the massive investments of a data center. Furthermore, there is no need to cool the infrastructure as the heat is directly used for heating purposes. Seasonality has already been taken into account by using CPU low power mode, install calculation units in strategic hosts such as schools that are closed during summer. In addition, computing power demand is seasonal, there is less jobs to perform during summer. Qarnot computing grid will be able to guarantee all year long a certain capacity, the extra power will be offered for free to universities and will be used as an adjustment variable (to meet the needs for calculation and the needs for heat). Qarnot can therefore propose a very energy efficient cloud HPC for reasonable prices. This is one key to address the HPC market which already represents billions of euros. Qarnot will be able to propose greener cloud HPC for lower prices than the competitors.

Qarnot is an innovative SME, gathering sixty high skilled people, located in France and acting at the European level. Thanks to the EIC-Accelerator, Qarnot aims at being the champion of the green cloud providers in Europe, a booming market with a double digit growth, in demand for sustainable technologies. To support its ambition, Qarnot relies on proprietary innovative technological bricks, both software and hardware. Qarnot IT infrastructures are natively designed to reuse waste heat, easily and efficiently, offering high performance computing and low environmental impact at the same time. Based on its successes and prestigious customers, Qarnot intends to promote a paradigm shift in the cloud services industry, by bridging performance and impact, security and European sovereignty, by considering IT as an energetic topic. With Qarnot, the more projects you install, the more impact you have: EIC will help us scaling, take an industrial turn and reach our high business ambitions.

Billion of connected devices are announced in 2020. This could cause a major revolution in ambiant intelligence if we can formulate appropriate architectures to process the massive data that will be produced or required by these devices. These last years, the question of the most adequate architecture for intelligent systems based on connected devices was intensively studied. Most of the proposed solutions were based on centralized models. Here, connected devices send data to a central platform (typically a cloud) where are implemented the services that process the data. Once the processing is done, results are sent, if needed, to the referred connected device. There exists several applications where the centralized models were successfully applied. Despite these successes, Internet of Things experts (Gartner, IBM etc.) say that these solutions are not sustanaible. Indeed, it is hard to manage network contention, confidentiality or real-time processing in centralized architectures. One of the most promising alternatives to these models consist of realizing the processing required by a set of connected devices by the devices themselves. The challenge then is to coordinate the set of devices in an environment for the realization of the required computations. Doing so, we build what we call cloud of things. Such systems are nowadays possible because of the increasing computing power of connected devices. In addition, clouds of things have interesting advantages on confidentiality, reactivity and energy consumption. The goal of the GRECO project is to develop a reference resource manager for cloud of things. The manager should act at the IaaS, PaaS and SaaS layer of the cloud. One of the principal challenges here will consist in handling the execution context of the environment in which the cloud of things operate. Indeed, unlike classical resource managers, connected devices imply to consider new types of networks, execution supports, sensors and new constraints like human interactions. The great mobility and variability of these contexts complexify the modeling of the quality of service. To face this challenge, we intend to innovate in designing scheduling and data management systems that will use machine learning techniques to automatically adapt their behavior to the execution context. Adaptation here requires a modeling of the recurrent cloud of things usages, the modeling of the physical cloud architecture and its dynamic. The GRECO project is built upon a collaboration between an enterprise (Qarnot Computing) and two French research institutes: the «Laboratoire d'Informatique de Grenoble (LIG)» and the «Institut National de recherche en informatique et automatique (Inria)». In the project, the LIG will bring its expertise in the design of schedulers for large systems. Inria will contribute in the design of a data management system for massive data. Qarnot will offer the expertise it gained in designing a resource manager for its network of digital heaters. This network will also be used for the validation of the project. However, the proposed solutions should be interoperable: they must be usable to build other systems like Edge/Extreme edge computing systems.

Data Centres (DC) are among the largest yet increasing energy consumers, due to the rising digitization of human activities. RES integration and energy efficiency improvements have the potential to reduce significantly DC carbon footprint, while increasing at the same time the auto-consumed energy, thus improving DC security and resiliency of supply against climate change. However, very few solutions, despite validated in lab, have been successfully deployed on operational DCs, mostly due to technological fragmentation, excessive CAPEX and lack of appropriate business models. CATALYST will address these challenges through turning existing/new DCs into flexible multi-energy hubs, which can sustain investments in RES and energy efficiency by offering mutualized flexibility services to the smart energy grids (both electricity and heat grids). By leveraging on the outcomes of FP7 GEYSER and DOLPHIN projects, CATALYST will adapt, scale up, validate and deploy an innovative, adaptable technological and business framework aimed at i) exploiting available DC non-IT legacy assets (onsite RES/backup generation, UPS/batteries, cooling system thermal inertia, heat pump for waste heat reuse) to deliver simultaneous energy flexibility services to multi-energy coupled electricity/heat/IT load marketplaces ii) deploying Cross-DC cross-infrastructures (e.g. heat vs IT) IT workload orchestration, by combining heat-demand driven HPC geographical workload balancing, with traceable ICT-load migration between federated DCs to match IT demands with time-varying on-site RES (“follow the energy approach”) iii) providing marketplace-as-a-service tools to nurture novel ESCO2.0 business models. The adaptation and replication potential of CATALYST will be demonstrated through carrying out four different real life trials spanning through the full spectrum of DCs types (fully distributed DCs, HPC, co-location, legacy) and architectures (from large centralized versus decentralized micro-DCs).