This FIL project in collaboration with Lyon2 University explores the design of machine learning algorithms for predicting energy consumption of large scale clusters of machines.
Avalon members : Laurent Lefevre and Jean-Christophe Mignot + Didier Puzenat (LIRIS, Univ. Lyon2)
Digital Infrastructures as the future Internet, constitutes the cornerstone of the digital transformation of our society. As such, Innovation in this domain represents an industrial need, a sovereignty concern and a security threat. Without Digital Infrastructure, none of the advanced services envisaged for our society is feasible. They are both highly sophisticated and diverse physical systems but at the same time, they form even more complex, evolving and massive virtual systems. Their design, deployment and operation are critical. In order to research and master Digital infrastructures, the research community needs to address significant challenges regarding their efficiency, trust, availability, reliability, range, end-to-end latency, security and privacy. Although some important work has been done on these topics, the stringent need for a scientific instrument, a test platform to support the research in this domain is an urgent concern. SLICES ambitions to provide a European-wide test-platform, providing advanced compute, storage and network components, interconnected by dedicated high-speed links. This will be the main experimental collaborative instrument for researchers at the European level, to explore and push further, the envelope of the future Internet. A strong, although fragmented expertise, exists in Europe and could be leveraged to build it. SLICES is our answer to this need. It is ambitious, practical but overall timely and necessary. The main objective of SLICES-DS is to adequately design SLICES in order to strengthen research excellence and innovation capacity of European researchers and scientists in the design and operation of Digital Infrastructures. The SLICES Design study will build upon the experience of the existing core group of partners, to prepare in details the conceptual and technical design of the new leading edge SLICES-RI for the next phases of the RI’s lifecycle.
PRACE, the Partnership for Advanced Computing is the permanent pan-European High Performance Computing service providing world-class systems for world-class science. Systems at the highest performance level (Tier-0) are deployed by Germany, France, Italy, Spain and Switzerland, providing researchers with more than 17 billion core hours of compute time. HPC experts from 25 member states enabled users from academia and industry to ascertain leadership and remain competitive in the Global Race. Currently PRACE is finalizing the transition to PRACE 2, the successor of the initial five year period. The objectives of PRACE-6IP are to build on and seamlessly continue the successes of PRACE and start new innovative and collaborative activities proposed by the consortium. These include: assisting the development of PRACE 2; strengthening the internationally recognised PRACE brand; continuing and extend advanced training which so far provided more than 36 400 person·training days; preparing strategies and best practices towards Exascale computing, work on forward-looking SW solutions; coordinating and enhancing the operation of the multi-tier HPC systems and services; and supporting users to exploit massively parallel systems and novel architectures. A high level Service Catalogue is provided. The proven project structure will be used to achieve each of the objectives in 7 dedicated work packages. The activities are designed to increase Europe’s research and innovation potential especially through: seamless and efficient Tier-0 services and a pan-European HPC ecosystem including national capabilities; promoting take-up by industry and new communities and special offers to SMEs; assistance to PRACE 2 development; proposing strategies for deployment of leadership systems; collaborating with the ETP4HPC, CoEs and other European and international organisations on future architectures, training, application support and policies. This will be monitored through a set of KPIs.
Optimizing resource management in a data center is crucial for economic and ecological reasons.
One of the key points is to distribute all virtual machines running on a minimum number of physical servers. With this in mind, we propose to study an optimization problem which is the migration of a set of virtual machines under different constraints in a datacenter. The aim is to determine the best migration sequence for a set of virtual machines from an initial state to a final state, which minimizes the total migration time, with or without intermediate migration. The problem can be modeled by a state graph.
Duration: 2019-2021
Project in cooperation with LIRIS.
Summary
Europe is undergoing a major transition in its energy generation and supply infrastructure. The urgent need to halt carbon dioxide emissions and prevent dangerous global temperature rises has received renewed impetus following the unprecedented international commitment to enforcing the 2016 Paris Agreement on climate change. Rapid adoption of solar and wind power generation by several EU countries has demonstrated that renewable energy can competitively supply significant fractions of local energy needs in favourable conditions. These and other factors have combined to create a set of irresistible environmental, economic and health incentives to phase out power generation by fossil fuels in favour of decarbonised, distributed energy sources. While the potential of renewables can no longer be questioned, ensuring reliability in the absence of constant conventionally powered baseload capacity is still a major challenge.
The EoCoE-II project will build on its unique, established role at the crossroads of HPC and renewable energy to accelerate the adoption of production, storage and distribution of clean electricity. How will we achieve this? In its proof-of-principle phase, the EoCoE consortium developed a comprehensive, structured support pathway for enhancing the HPC capability of energy-oriented numerical models, from simple entry-level parallelism to fully-fledged exascale readiness. At the top end of this scale, promising applications from each energy domain have been selected to form the basis of 5 new Energy Science Challenges in the present successor project EoCoE-II that will be supported by 4 Technical Challenges
Partners
CEA, FZJ, ENEA, BSC, CNRS, INRIA, CERFACS, MPG, FRAUNHOFER, FAU, CNR, UNITN, PSNC, ULB, UBAH, CIEMAT, IFPEN, DDN, RWTH, UNITOV
Project Information
EoCoE-II is a H2020 RIA european project, call H2020-INFRAEDI-2018-1.
Duration: 3 years, Jan 1st 2019, Dec 31st 2021.
Avalon Members: T. Gautier, C. Perez
Online Resources
L’objectif du projet FENNEC de l’ANR-MRSEI est de faciliter le montage d’un projet pour l’appel DT-NMBP-08-2019 intitulé « Real-time nano-characterisation technologies (RIA) ». La caractérisation à l’échelle nanométrique permet d’avoir des informations uniques sur la structuration et les propriétés des matériaux et des dispositifs mais nécessite une expérience pointue dépendant des matériaux envisagés, des temps d’acquisition, de dépouillement et d’analyse, incompatibles avec les contraintes d’une ligne de production industrielle. Cet appel souhaite réduire d’abord ces différentes contraintes notamment en accélérant l’acquisition et l’analyse des données instrumentales. Il veut aussi pouvoir valoriser plus facilement les résultats de nano-caractérisations avancées.
Start Date: 2018, August 1st
Duration: 24 months
Leader: T. Deutsch (CEA/MEM)
Avalon Members: C. Perez, E. Caron
Associated joint team between Avalon and RDI2 Lab. in Rutgers University on Sustainable Ultra Scale compuTing, dAta and energy Management
The collaboration proposed by the SUSTAM associate team focuses on aspects of sustainability in ultra-scale systems. Launched in 2017, the SUSTAM associated team aims to design a multi-criteria orchestration framework that manages resources, data and energy consumption in an efficient manner. The SUSTAM associate team will enable a long-term collaboration between the Inria Avalon and the RDI² team (Rutgers University . It will allow the teams to coordinate efforts and pursue common research activities in topics such as sustainable software solutions, resource and big-data management, elasticity of stream and batch applications, and energy efficiency. The involved members will contribute to the design of a common architecture and framework with components and algorithms adapted to various contexts.
Web Site: http://avalon.ens-lyon.fr/sustam/
The goal of PMSISEE is to support the collaboration between the Avalon (LIP) and Beagle (LIRIS) teams through research activities on programming modelsand tools for HPC applied to the Aevol/R-Aevol simulator of in silico evolution of bacteria.
A population of organisms adapting to a new environment is a dynamic system changing over time at many levels (molecules, networks, individuals, ecosystems). A large amount of empirical and theoretical evidence indicates that in real populations all these levels interact, making the dynamics of adaptation a highly complex phenomenon. In order to understand bacterial evolution, we need large-scale integrative models in which all relevant levels from the molecule to the ecology are simulated. The Aevol/R-Aevol simulator (http://www.aevol.fr) has been developed by the Beagle team to address such questions. Aevol integrates the molecular and cellular levels to address the evolution of genomic complexity. R-Aevol adds the network level to investigate the evolution of network complexity.
In this project we consider the Aevol/R-Aevol simulator, or equivalent code, as the object of the study. At a first glance, it is characterized by several properties: the code is complex due the models to integrate; the amount of computational resources required for simulations is huge when considering the size of the systems (millions of base pairs in the genome, thousands of genes in the genetic network, billions of individuals in the population, billions of generations); load unbalance occurs when running the models under different conditions ( i.e., different parameters). Any gain in performance, will make these simulations very valuable to understand bacterial evolution and to have feedback on the biological models in order to improve them.
The research during the PMSISEE project will be restricted to two main issues related to the software and its algorithms: 1/ analysis and design of specialized models to tackle software complexity in the context of HPC using next generation of parallel supercomputers. This point is based on advances in software engineering of these last twenty years in particular with respect to code composability and re-use using component model; 2/ performance analysis and design of new, or improvement of existing, algorithms for scalable and efficient simulation of evolving bacterial populations on modern parallel architecture. This axis will deal with heuristics for scheduling in order to well balance the work load and reducing communication.
Duration: 2017-2019
According to the current studies, the telecom infrastructure is the major contributor for the ever increasing energy demand in the ICT sector and has a major part on carbon footprint to the environment. And surprisingly, more than 80% of this share is consumed by the Home Gateways (HGs).
Hence, in this preliminary work, we have explored the possibility of relocating some of the functionalities of a HG into a vHGW (virtual Home Gateway) which is hosted by a node located in NSP premises. Based on our experiment, it was possible to host up to 1000 vHGWs on a single server machine which consumes around 100W. And our result showed that the number of vHGWs hosted on server machine does not have a significant variation on its energy consumption. We have also confirmed that the capability of a vHGW’s in the provision of the network and application level services such as, routing, DHCP, firewalling and NAT, alike HG’s.
If we consider a replacement of the current HG by a quasi passive device (which can consume around 1Watt) and if we suppose that end users have triple play services over a fiber link (FTTH). By pulling those network and application level services into a vHGW and using a server machine that can host around a 1000 vHGW’s (and probably more in a near future), we can obtain about 300% energy saving in the overall wire line telecom networks. Therefore, the result of our experiment is aligned to and complies with the recommendation set by the GreenTouch project (http://greentouch.org).
Hence, the result of this study shows the benefit of service relocation of HG’s by reducing significantly the overall energy consumption of a wire line network, and minimizing the sector’s impact on the environment.
For more information about this research work. please visit vHGW Web page.
To accommodate the ever-increasing demand for Utility Computing (UC) resources, while taking into account both energy and economical issues, the current trend consists in building larger and larger data centers in a few strategic locations. Although such an approach enables UC providers to cope with the actual demand while continuing to operate UC resources through centralized software system, it is far from delivering sustainable and efficient UC infrastructures. We claim that a disruptive change in UC infrastructures is required: UC resources should be managed differently, considering locality as a primary concern. To this aim, we propose to leverage any facilities available through the Internet in order to deliver widely distributed UC platforms that can better match the geographical dispersal of users as well as the unending demand. Critical to the emergence of such locality-based UC (LUC) platforms is the availability of appropriate operating mechanisms. We advocate the implementation of a unified system driving the use of resources at an unprecedented scale by turning a complex and diverse infrastructure into a collection of abstracted computing facilities that is both easy to operate and reliable.
Start Date: January 2015
Duration: 4 years
Avalon Members: J. Darrous, G. Fedak, C. Perez
More information on Discovery website
The DISCOVERY initiative aims at exploring a new way of operating Utility Computing (UC) resources.