Inria - Software assets - Energy
Upcoming SlideShare
Loading in...5
×
 

Inria - Software assets - Energy

on

  • 1,014 views

Inria's software assets Cities, Energy and Climate for France India Techno Summit october 2013

Inria's software assets Cities, Energy and Climate for France India Techno Summit october 2013

Statistics

Views

Total Views
1,014
Views on SlideShare
1,014
Embed Views
0

Actions

Likes
0
Downloads
13
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Inria - Software assets - Energy Inria - Software assets - Energy Presentation Transcript

    • Inria, a public research institute established in 1967, produces excellent research in mathematics and computational sciences. The institute covers the entire spectrum of research at the heart of these areas of activity, and works also on issues asked by other sciences and by economic and societal actors. Inria can be considered as a software provider. Indeed, approximately 1000 software are currently developed of which a large part is distributed with an open source license. Inria is convinced that this production deserves to be more widely known among academic and economical actors as it is a source of value creation, which is one of Inria's missions. The India France Technology Summit is an opportunity to raise awareness of this software production. Thus, this catalog offers an extract from software assets, as part of the Summit themes such as health, smart cities, smart grid, critical embedded software, etc. . You can find all these software files, and many others on Icommunity: https://www.icommunity.fr/patrimoine-technologique-inria
    • Contact: sylvain.karpf@inria.fr http://www.apisense.fr APISENSE® APISENSE platform is a software solution to collect various contextual information from Android devices (client application) The software architecture of APISENSE ensures privacy to the mobile application users (CNIL certified) and preserves the device’s battery. and automatically upload collected data to a server (deployed as a SaaS). APISENSE computing datasets is based on infrastructure collection from to a Cloud facilitate significant populations of mobile users for research purposes. Due to its openness, the platform is adaptable to different requirements and fosters the deployment of experiments. Language & environment : Java, Scala, JavaScript & Python available for Windows, Unix and MacOS X. Android mobile application Mots clefs : mobile sensing, crowd-sensing, crowd-sourcing, Android Licence : open platform for research activities and proprietary license for industrial applications Academic partner:
    • Contact: chantal.le_tonqueze@inria.fr http://www.eclipse.org/m2m/atl/ ATL (AtlanMod Transformation Language) is a Innovative aspects: Support for all model types (e.g., UML2), tool dedicated to model transformation, declarative (QVT-like) and imperative complete, integrated into Eclipse and including syntaxes. its own development & execution environment, Main fields of application: concrete examples as well as the Code generation preparing, reverse corresponding associated documentation engineering of legacy systems, tool interoperability, etc. More generally, any context implying model processing. Keywords: MDA, MDE, ATL, QVT, model, transformation Language, standard, environment: Eclipse, EMF, Java License: Eclipse Public License (EPL) Partners:
    • Contact: Gabriel.Antoniu@inria.fr http://blobseer.gforge.inria.fr/ Innovative aspects: BlobSeer leverages BlobSeer is a large-scale distributed storage service for advanced management of massive versioning and distributed metadata management to sustain a high throughput data. under heavy data access concurrency. Validated on Nimbus, OpenNebula and Up to 30% improvement for MapReduce Microsoft Azure cloud platforms. applications with respect to Hadoop. Main fields of application: Advanced storage services for data-intensive cloud applications. Language : C++, Python, Ruby License : LGPL Keywords: cloud storage, versioning, distributed metadata, MapReduce Academic partners:
    • Contact: pierre.martinon@inria.fr URL: www.bocop.org controlled Innovative aspects: highly flexible toolbox, with state-of-the-art algorithms, advanced and user-friendly GUI Features powerful solvers and user-friendly GUI. Fit for both academic and industrial use. Main fields of application: transportation, Toolbox for optimization dynamical systems. of energy, process optimization, and biology Language, environment: C++, Linux / Mac / Windows Keywords: control, optimization, toolbox License: EPL (core) Possible proprietary modules Academic partners :
    • Contact: ALPINES Laura.Grigori@inria.fr https://who.rocq.inria.fr/Laura.Grigori/ CALU solves linear systems of equations Ax=b Innovative aspects: CALU addresses a major challenge in HPC, using Communication Avoiding LU. the exponentially increasing cost of Static scheduling communication with respect to computation. It attains lower bounds on communication and it employs novel lightweight scheduling Static + 10% dynamic scheduling techniques to be able to adapt to dynamic changes in the system. Progressively different algorithms as QR, 100% dynamic scheduling RRQR, will be integrated in the library. time Main fields of application: can be used to solve very ill-conditioned problems or as a building block of iterative methods. Programming language: C and Pthreads Keywords: linear algebra, communication avoiding License : open source
    • Contact: sriv-saclay@inria.fr http://coco.gforge.inria.fr/doku.php Choose your Optimizer! COmparing Continuous Optimizers is a platform for benchmarking numerical optimization algorithms in order to determine the best optimizer for a given class of problem. Innovative aspects: Tedious task of running experiments, postprocessing and visualization entirely automatized. Database of different benchmarked algorithms with the platform available. Main fields of application: Any field where one needs to solve numerical optimization problem in black-box scenario, such as biology, physics (for parameter calibration), engineering design (launcher, airplane, car, ...). Language, environment: C, Java, Matlab and Octave, Python Keywords: benchmarking, numerical optimization, black-box optimization. License: BSD Academic partners:
    • http://wimmics.inria.fr/corese Corese is a Semantic Web Factory implementing W3C RDF, RDFS, SPARQL 1 .1 Query & Update as well as SPARQL rules for RDF Innovative aspects: Corese 3.0 is a lightweight generic modular implementation of RDF(S) and SPARQL 1.1 in main memory Main fields of application: Generic Engine e-learning, biology, geology, engineering design, cartography, technological watch, social networks analysis, etc Language, environment: Java : Linux, MacOS & Windows Keywords: Semantic Web, Social Web, Search Engine, RDF, SPARQL License: Open source (CeCILL-C) Academic partners: I3S, CSTB, INRA, IFP, IGN, … Services offering & developments :
    • Contacts: julien.bruneau@inria.fr charles.consel@inria.fr http://diabox.inria.fr DiaSuiteBOX proposes an application store that gathers the devices deployed at home. This store is open and available online such as an application store for Smartphone. Innovative aspects: - Creation of innovative applications that enable cross-device interoperability - Customizable and evolutive smart home platform. - Reduced cost of our platform by using and gathering inexpensive devices. Main fields of application: Smart Home, Assisted Living Language, environment: Java, Eclipse Keywords: Smart home, Open application store, Development tool suite, Application certification. License: A previous version of our DiaSuite tool suite is available online: http://diasuite.inria.fr.
    • Graphite Contact: catherine.lucquiaud@inria.fr http://alice.loria.fr/software/graphite Innovative aspects: Graphite offers a wide range of functionalities numerical geometry that enables, among other including parameterisation, multi-resolution things, 3D modelling and texture baking. analysis and digital optimisation. Graphite is a dedicated software platform in Graphite was awarded two prizes at the "Trophées du Logiciel Libre 2009", a special prize for the "most innovative project" and third prize in the "scientific software" category. Main fields of application: all graphic applications (CAD, video games, etc) Language/System: C++ language, Available for Linux and Windows. Keywords: CAD, 3D modelling, Numerical Geometry, Visualisation, Texturing, Lighting. License: GNU GPL
    • H2OLab Contact: jocelyne.erhel@inria.fr URL: http://h2olab.inria.fr/ Innovative aspects:  Modeling and numerical simulation of aquifers  Generation of random fields  Porous and fractured heterogeneous media  Flow with mixed finite elements  Solute transport with a Lagrangian method  Stochastic modeling for data uncertainty. Main fields of application:  Generation and meshing of random fracture networks  Multilevel parallelism : subdomains, particles, random samples.  Groundwater resources management  Prevention and remediation of pollutions  Nuclear waste deep storage  CO2 sequestration  Energy resources exploitation (geothermy, oil, gas). Language, environment: C++ Modular structure, distributed parallelism with MPI Integrated development environment (Visual, Eclipse) Continuous integration server (Jenkins) Project management (Gforge, Redmine) Free numerical libraries (CGAL, HYPRE, etc). Keywords: Modeling, simulation, multiscale, uncertainly, heterogeneity, hydrogeology, groundwater, contamination, energy. License: license agreement on demand MICAS consortium of 4 research laboratories. Academic partners:
    • Contact: pascal.moussier@inria.fr http://hips.gforge.inria.fr/ Hips is a hybrid “à la carte” software programme for solving sparse linear equation systems with the degree of precision and time/memory commitment required and desired by the user. Innovative aspects: Hips is THE solution for solving systems combining several difficulties: very large n (number of equations in the system >1 billion), 3D mesh and high level of complexity. Hips is hybrid: according to the needs of the problem, Hips mobilises direct or iterative solving methods. Main fields of application: - simulation of physical phenomena (reservoirs, electromagnetics) - magnetohydrodynamics - fluid flows - manipulation of abstract concepts (finance...) Language: Fortran, C Partition of space around an airplane for parallel computing. Solving a system with Hips: Using the iterative method with interfaces Using the direct method inside domains Keywords: HPC, parallel computing, solving sparse linear systems, hybrid direct iterative method. License : LGPL/CECILL C
    • http://hop.inria.fr/ HOP Programming language which uses the web as an execution platform and targets interactive multimedia applications Innovative aspects: HOP offers, in a single syntax, all of the functionalities necessary for the programming of multimedia applications, from communication protocols (http) to traditional algorithmics as well as the generation of HTML pages. Language, environment: - C (Linux, MacOS, Windows, Android) - Java virtual machine Keywords: Web 2.0, house automation, multimédia, office License : Open source (GPL)
    • Contact: Brice.Goglin@inria.fr http://www.open-mpi.org/projects/hwloc/ « Portable Hardware Locality » hwloc discovers your computers’ topology and exposes it as an abstracted and portable hierarchical tree. Main fields of application: hwloc’s main goal consists in assisting applications for retrieving easy-to-use information about modern compute server architectures so as to exploit their power in a more convenient and efficient manner. Language: C Keywords: Topology, locality, affinities, HPC, MPI, GPU License: new BSD licence Technological barrier: Users may now visualize the topology within a graphical interface and consult many architectural and computing characteristics. Academic partners: Open MPI consortium, Argonne Nationa Lab., etc.
    • Java In The Small JITS –Java In The Small– is a software platform which allows Java applications to be migrated automatically to various embedded systems such as smart cards, networked sensors or minimalist games consoles. Contact: sylvain.karpf@inria.fr http://stan-project.gforge.inria.fr Innovative aspects: The JITS tool chain can be used to roll out software on very restricted platforms. It is therefore possible to run Java programs on media for which a conventional virtual machine does not work. JITS also incorporates analysis software including STAN, which, during roll-out, checks that there are no data leakages where the software is installed and thus imposes a high security level. Keywords: Embedded systems, Java, highly restricted environments. License: pperating licences are negotiated on a case by case basis. Academic partners:
    • Contact: sriv-saclay@inria.fr http://gforge.inria.fr/projects/kadeploy3 Kadeploy 3 is a full rewrite of the Kadeploy 2 tool initially developed at IMAG. In order to achieve a larger scale, the design was completely reviewed and Kadeploy 3 is based on dedicated large-scale tools. Since the deployment process operates at a low level, special care was devoted to the detection of critical incidents in the deployment of large numbers of nodes. Innovative aspects: Kadeploy 3 offers users the flexibility required to personalise deployment and thus meet all their reconfiguration needs. Main fields of application: Grid'5000 platform, academic computing centres and industrial cluster users Language, standard, environment: Ruby, Bash Keywords: environment configuration License: CECILL Licence V2
    • Contact: chantal.le_tonqueze@inria.fr http://kermeta.org/ Innovative aspects: programming environment based on an object- Kermeta allows to define and enrich new programming languages by improving the way oriented DSL (Domain Specific Language) to specify, simulate and test operational semantics of metamodels. optimized for metamodel engineering. The Kermeta workbench is a powerful meta- . Main fields of application: Agile software developent, model engineering Key words : MOF (Meta Object Facility), OMG (Object Management Group), Metamodelisation Coding and Operating System: Java, Kermeta License: EPL Academic Partners:
    • Contact: christophe.biernacki@inria.fr http://www.mixmod.org Mixmod is a software package for ModelBased supervised and unsupervised classification on qualitative and quantitative data. Innovative aspects: Mixmod included a library of calculation (mixmodLib) and tools adapted to a wide spectrum of users and needs: a graphical user interface (mixmodGUI), functions for R environments (Rmixmod) and Matlab (mixmodForMatlab).. Language: C++, R, Matlab et Scilab Environment: Windows & Linux Keywords: classification , mixed data, data mining, data modeling. License: GNU GPL Academic partners:
    • Contact: sara.hachem@inria.fr MobIoT Middleware for the Large-scale Internet of Mobile Things that handles the large number of active mobile devices by preventing redundant ones from registering their services Innovative aspects: Support the development of IoT applications based on knowledge about the real world. Handle a large number (billions) of mobile devices based on probabilistic models 1 2 Lookup 3 Registry Registration Access 2 Main fields of application: participartory mobile sensing , IoT applications Language, environment: Java, Android Keywords: mobile Internet of Things 1 License: LGPL, Dual License 3 Service Consumer Service Providers Academic partners
    • Contact: sylvain.karpf@inria.fr http://www.moosetechnology.org Moose is an extensive platform for software and data analysis. It offers multiple services ranging from importing and parsing data, to modeling, to measuring, querying, mining, and to building interactive and visual analysis tools. Innovative aspects: Moose accelerates production of software analyses by providing expandable models and reusable libraries. Several models are offered for static or dynamic analysis of programmes, development logs, procedural code or aspect. Smalltalk (open-source Pharo) Windows, Linux, Mac Main fields of application: Retroengineering, Meta-modelling, Software visualisation License: BSD licence Academic partners:
    • Contact: rev@inrialpes.fr http://graal.ens-lyon.fr/MUMPS Multifrontal Massively Parallel Solver solves sparse linear systems of the form A x = b. Innovative aspects: MUMPS is a direct solver, used worldwide in many fields related to digital simulation, in academic as well as industrial codes. Its main characteristics are its numerical stability, large number of functionalities and high performance. Language, standard, environment: Fortran and C languages, available for Unix, Windows or Mac OS, parallel calculators. Keywords: sparse matrices, direct solvers, highperformance calculation, finite element modelling Licence: Open source Spinning wheel (courtesy of SAMTECH) Academic partners:
    • Contact: regis.duvigneau@inria.fr http://num3sis.inria.fr - Visual programming (dataflow) - Coupling via plugins - Interactive computation and visualization - Distributed computing Innovative aspects: Plateform for the aggregation of software components (plugins) dedicated to multidisciplinary numerical simulation in HPC environment Coupling of models (aerodynamics, electromagnetics, structural mechanics), coupling of numerical methods (finitevolume, finite-element, discontinuous Galerkin, linear solver, mesh generation, etc) Language, environment: C++, libraries Qt, vtk, mpi Keywords: simulation, plateform, HPC License: free for academic usage
    • Contact: rev@inrialpes.fr http://oar.imag.fr A task and resource management tool for computing infrastructures. Innovative aspects: Versatility, customisation, scaling up, control of complexity, flexible scheduler. Main fields of application: Use of infrastructures for high-performance computing. Language: Perl, Ruby, Ocaml, Bash, SQL. Available for UNIX: Debian, Centos, RedHat. Keywords: resource manager, clusters, light grid, cloud, HPC. License: GPL Partners: Used by (among others) Grid’5000, SensLAB, Mesocentre CIMENT, BRGM.
    • OCaml The OCaml language is a functional programming language that combines safety with expressiveness through the use of a precise and flexible type system with automatic type inference. The OCaml system is a comprehensive implementation of this language, featuring two compilers (a bytecode compiler, for fast prototyping and interactive use, and a native-code compiler producing efficient machine code for x86, ARM, PowerPC and SPARC), a debugger, a documentation generator, a compilation manager, a package manager, and many libraries contributed by the user community. Contact: Xavier.Leroy@inria.fr http://caml.inria.fr/ http://ocaml.org/ Innovative aspects: OCaml is one of the three most widely-used functional programming languages, and the only one that seamlessly combines functional, imperative and object-oriented programming within a single type system. Main fields of application: Symbolic computation : compilers, code generators, static analyzers, formal verification tools, automated theorem provers, interactive proof assistants. High-level systems, network and Web programming. OCaml is also used in the financial industry, notably for modeling and pricing complex financial products. Language, environment: Programmed in OCaml and in C, the OCaml system supports Linux, BSD, MacOS X, and Microsoft Windows. Keywords: functional programming, static typing, symbolic computation, compilation, static analysis., License: OCaml is free software, distributed under QPL and LGPL licences. A specific BSD-like license is offered to members of the Caml Consortium.
    • Contact: chantal.le_tonqueze@inria.fr http://www.openmask.org/ OpenMASK (Modular Animation and Simulation Kit) is a platform for modular heterogeneous and distributed applications supporting development and execution in animation, simulation and virtual reality fields. Innovative aspects: In OpenMASK, the modularity unit is the simulated object (OSO). It can be used to model and to control an entity or a virtual object behaviour or motion control as well as imput devices. Main fields of application: Manufacturing industry, sports, education, training. Key Words: Virtual reality, simulation, manmachine interfaces. Coding and Operating System: C++, Windows and Linux. License: QPL Academic Partners:
    • Contact: sylvain.karpf@inria.fr http://paradiseo.gforge.inria.fr/ ParadisEO Paradiseo is a software framework for metaheuristics (optimisation algorithms aimed at solving difficult optimisation problems). It facilitates the use, development and comparison of classic, multi-objective, parallel or hybrid metaheuristics. ParadisEO-PEO ParadisEO-MO Innovative aspects: ParadisEO offers implementations of numerous classic metaheuristics from literature as well as innovative new metaheuristics. This software can also be used for parallelisation of metaheuristics. Main fields of application: ParadisEO is used to solve large-scale optimisation problems in a wide variety of fields, such as transport, logistics, telecommunication systems and bioinformatics. ParadisEO-MOEO ParadisEO-EO EO : Population-based metaheuristics Language, environment: available for Windows, Unix and Mac OS X, Language C++, CMake Keywords: Metaheuristics, parallelisation. MO : Single solution-based metaheuristics MOEO : Metaheuristics for multiobjective optimisation PEO : Hybrid, parallel and distributed metaheuristics License: freeware available Academic partners:
    • Contact: sylvain.karpf@inria.fr http://www.pharo-project.org Pharo By providing a stable and small core system, excellent dev tools, and maintained releases, Pharo is an attractive platform to build and deploy mission critical Smalltalk applications. Innovative aspects: Providing a stable system, a simple kernel, excellent development tools and regular updates, Pharo is an attractive platform for creating and rolling out critical applications in Smalltalk. Language, environment: Operates under Windows, Unix, Mac OS X and iPhone. Keywords : Object programming, reflective system, Smalltalk. License: MIT Academic partners:
    • Contact: marianne.lamour@inria.fr http://www.labri.fr/~pelegrin/scotch/ (PT-)Scotch Scotch is a graph partitioner. It helps optimise the division of a problem, by means of a graph, into a set of independent sub-problems of equivalent sizes. These sub-problems can also be solved at the same time. Partitioning of edges into 8 parts. Each part is allocated to a different processor. Partitioning vertices to reorder blocks of a sparse matrix. Innovative aspects: Scotch calculates partitions on a sequential machine and PT-Scotch on a parallel machine. PT-Scotch can handle very large graphs (2 billion vertices or more) on several tens of thousands of processors. The quality of solutions provided by PT-Scotch does not depend on the number of processors on which it runs. Main fields of application : distributing a scientific problem's computing load across the processors of a parallel machine, placement of VLSI circuits, combinatorial optimisation, etc... Language: C Keywords: high-performance calculation, graph algorithmics, domain decomposition, mesh partitioning, sparse matrix ordering License: CECILL C
    • proactive.inria.fr ProActive Portable Java middleware designed for the distributed and parallel multi-core calculation. It allows the progamming and the execution of parallel applications as well as ressources management (workstations, servers, clusters, grids or cloud) Innovative aspects: Distributed active objects, seamless components migration, large scale communicating software components Main fields of application: Distributed systems and services Language, environment: Java (Linux, MacOS, Windows) Keywords: distributed computing, grids, cloudcomputing License : - Open source (AGPL) - Proprietary within
    • Contact: rev@inrialpes.fr http://nano-d.inrialpes.fr/ SAMSON A software platform for real-time modelling and Innovative aspects: simulation of natural or artificial nanosystems. The specific interest of SAMSON lies in the integration of simulation methods during the modelling phase: adaptive and interactive simulation algorithms immediately provide information on the consequences of modelling choices. Language, standard, environment: C++, available for Windows, Unix and Mac OS X Keywords: Modelling, Simulation, Nanosystems, Structural Biology, Chemistry Academic partners
    • Contact: bertrand.thirion@inria.fr http://scikit-learn.org Scikit-learn is a Python module integrating classic machine learning algorithms in the tightly-knit scientific Python world. It aims to provide simple and efficient solutions to learning problems, accessible to everybody and reusable in various contexts: machinelearning as a versatile tool for science and engineering. Innovative aspects: the added value of Scikit-Learn lies with the great quality and robustness of the functions available in its library. All functions are also extremely well documented. Main fields of application: Data mining, and more generaly any field requiring robust machine-learning solutions Language, environment: Scientific Python (Scipy) Keywords: Machine-Learning, Classification, regession , clustering License: BSD Academic partners: Telecom Paris, CEA Industrial Partners: Nuxeo, Saint Gobain, Logilab, Tinyclues
    • Selalib Selalib (SemiLagrangian Library) is a collection of modules conceived to aid in the development of plasma physics simulations, particularly in the study of turbulence in fusion plasmas. Selalib offers basic capabilities from general and mathematical utilities and modules to aid in parallelization, up to pre-packaged simulations. Contact: chacongolcher@math.unistra.fr http://selalib.gforge.inria.fr/ Innovative aspects: unavailability of basic software building blocks for simulations using the semilagrangian method. Need to include software engineering practices in an academic research environment. Main fields of application: high-level modules in the library are specialized for plasma physics. Low-levels are usable in parallel computing applications. Language, environment: Fortran 2003 Mettre une image représentative des fonctions du logiciel Keywords: plasma physics, semilagrangian method, parallel computing, plasma turbulence License: open source (Cecill-B), not released yet. Academic partners: University of Strasbourg, Max Planck Insitute - Garching.
    • Contact: catherine.lucquiaud@inria.fr http://simgrid.gforge.inria.fr/ Scientific Instrument for the study of Large-Scale Distributed Systems SimGrid is Easy-to-use Users can quickly setup a controlled simulation environment to develop and test new ideas with minimal technical effort. SimGrid is Versatile It provides models of Network, CPU and Disks adapted to (Data)Grids, P2P, Clouds, Clusters and HPC, allowing multi-domain studies. SimGrid ⇝ + Idea or MPI code Models Experimental Setup SimGrid is Trustworthy The models are theoretically grounded and experimentally validated. The results are reproducible, enabling better scientific practices. Scientific Results SimGrid is Scalable The simulation engine uses algorithmic and implementation techniques toward the fast simulation of large systems on a single machine. SimGrid is Open Due to its community-development philosophy, SimGrid is Usable it has been adopted by many researchers around the globe. Several associated tools are Ported to all major platforms and programming languages, it comes with an extensive tooling. contributed by its large user community. Keywords Large-Scale Distributed Systems; Reproducible Experiments; Simulation, Emulation. License: GNU LGPL Academic Partners:
    • Contact: christine.morin@inria.fr http://snooze.gforge.inria.fr/ Snooze : a Self-Organizing and Energy Aware Innovative aspects: - Energy conservation in clouds Cloud Management Framework. - Generic scheduling engine supporting VM placement, relocation and reconfiguration algorithms - Fault-tolerance - Self-organizing self-healing hierarchical architecture. Main fields of application: - Private cloud providers - Testbed for evaluation of VM scheduling algorithms in a real environment Language, environment: Java Keywords: Cloud computing, energy management, consolidation, fault-tolerance, self-organization, self-healing, virtualization License: GPL v2
    • Raymond.namyst@inria.fr StarPu http://runtime.bordeaux.inria.fr/StarPU/ A runtime system for heterogeneous parallel machines Applications Parallel Compilers Parallel Libraries Main fields of application: • Energy, Weather Forecast, Aeronautics, Multimedia StarPU Drivers (CUDA, OpenCL) CPU GPU Goals: • Dynamically schedule tasks on all processing units - CPU + GPU, Xeon Phi • Minimize data transfers between accelerators Target for • Compilers: StarSs [UPC], HMPP [CAPS entreprise] • Parallel Libraries: MAGMA [UTK, USA] Xeon Phi Language: C (C++, Fortran) Keywords: GPU, multicore, scheduling, performance License: LGPL Academic partners:
    • SUP Contact: Francois.Bremond@inria.fr https://team.inria.fr/stars/software Scene Understanding Platform SUP is a software platform for perceiving, analyzing and interpreting a 3D dynamic scene observed through a network of sensors. It encompasses algorithms allowing for the modeling of interesting activities for users to enable their recognition in real-world applications requiring high-throughput. Innovative aspects: Detection of human (and vehicle or other) activities through video analysis and other sensors in big databases. Main fields of application: Health-care, security, protection of people and goods Language, environment: C++, cmake, multi-platform Keywords: activity recognition License: undetermined Academic partners: CHU Nice, CEA, USC California, Tainan (Taiwan), Hamburg University, I2R Singapore
    • Contact: verdandi-help@lists.gforge.inria.fr http://verdandi.gforge.inria.fr/ Verdandi is a C++ library for data assimilation. Innovative aspects: Verdandi shows a It includes several methods for the coupling great genericity which makes it adapted to between one or several numerical models and many numerical models (regardless of their observational data. Verdandi also includes language, data types, parallelization tools to ease the application of data strategy, ...). Computing performance is not assimilation, in particular in the management of impaired by this genericity. observations or for a priori uncertainty quantification. The library may be used with Main fields of application: The library may model implemented in Fortran, C, C++ or be used with any model, usually of high Python. dimension, to be coupled with observations, for state estimation or inverse modeling. Langage, standard, environnement: C++, Python interface. Compatibility with Linux, MacOS et Windows. Keywords: scientific computing, model, observation, data assimilation, state estimation, inverse modeling, uncertainties Licence: GNU LGPL, version 2.1 or later Nitrogen dioxide concentrations over Clermont-Ferrand, after the assimilation of observations (discs).
    • Contact: george.rosca@inria.fr http://yarta.gforge.inria.fr/ Middleware for managing social ecosystems which offers rich social interactions for mobile environments. Innovative aspects: advanced social access control mechanism, ubiquitous decentralized information system, interapplication data reusage, rich API for mobile specific social sensors Main fields of application: social application development, rich social interactions Language, environment: Java SE/Android Keywords: rdf, semantic web, social networks License: Dual, LGPL v2.1