Embedded Files
I work on efficient support for AI (Federated Learning in particular) execution on large scale infrastructures, reproducibility and Big Data management on the Computing Continuum (clouds, fog, edge), stream processing and workflow data management.
I work on efficient support for AI (Federated Learning in particular) execution on large scale infrastructures, reproducibility and Big Data management on the Computing Continuum (clouds, fog, edge), stream processing and workflow data management.
Software
Software
E2CLab is a framework that implements a rigorous methodology that allows to deploy real-life application workflows to representative settings of the physical infrastructure underlying this application in order to accurately reproduce its relevant behaviors and therefore understand end-to-end performance.
ProvLight is a framework that allows researchers to efficiently capture provenance data of workflows running on IoT/Edge infrastructures. ProvLight presents low capture overhead in terms of: capture time; CPU and memory usage; network usage; and power consumption.
Planner is middleware for cost-efficient execution plans placement for uniform stream analytics on Edge and Cloud. Planner automatically selects which parts of the execution graph will be executed at the Edge in order to minimize the network cost. Real-world micro-benchmarks show that Planner reduces the network usage by 40% and the makespan (end-to-end processing time) by 15% compared to state-of-the-art.
KerA is a low-latency storage for stream processing (currently under development at Inria, in collaboration with Universidad Politécnica de Madrid, in the framework of a contractual partnership between Inria and Huawei Munich). By eliminating storage redundancies between data ingestion and storage, preliminary experiments with KerA successfully demonstrated its capability to increase throughput for stream processing.
Tyr is a transactional object storage system aimed at storage-based convergence between HPC and Big Data. Tyr natively offers data access coordination in the form of transactions. It offers a POSIX-compliant, transactional, distributed file system implementation built as a thin layer atop. This file system performs well and shows near-linear scalability properties on both HPC and Big Data platforms.
JetStream is a high performance batch-based streaming middleware. The system is able to self-adapt to the streaming conditions by modeling and monitoring a set of context parameters. It further aggregates the available bandwidth by enabling multi-route streaming across cloud sites.
TomusBlobs is a concurrency-optimized data storage system which federates the virtual disks associated to the Virtual Machines running the application code on the cloud. It is used in the Azure cloud for efficient data-intensive processing based on MapReduce.
MonALISA, which stands for Monitoring Agents using a Large Integrated Services Architecture, has been developed over the last years by Caltech and its partners with the support of the U.S. CMS software and computing program. The framework is based on Dynamic Distributed Service Architecture and is able to provide complete monitoring, control and global optimization services for complex systems.
Projects
Projects
PEPR CLOUD Steel (2023-2028): I am the WP2 leader of this national project focusing on efficient and secure data storage and processing on cloud-based infrastructures. Budget: 2.8M Euros.
Inria-DFKI Engage (2022-2025): I am the WP2 leader of this collaborative project between Inria and DFKI focusing on faster and more reliable AI in the processing of complex tasks. Budget: 500K Euros.
EuroHPC ACROSS (2021-2024): I am one of the Inria technical correspondants of this EuroHPC project focusing on combining traditional HPC techniques with Artificial Intelligence and Big Data analytics to enhance application outcomes. Budget: 4M Euros.
PHC PROCOPE FlexStream (2020-2022): I was the Principal Investigator of this collaborative project with University of Dusseldorf focusing on automatic scaling of streaming applications. Budget: 20K Euros.
ANR OverFlow (2015-2021): I was the Principal Investigator of this ANR JCJC project focusing on Workflow Data Management as a Service for Multi-site Applications. Budget: 250K Euros.
Inria Project Lab – HPC Big Data (2018-2022): I worked with other Inria teams at the intersection of HPC, Big Data and IA. In this context, I supervised the thesis of Daniel Rosendo.
BigStorage (2015-2018) is an European Training Network (ETN) project. Area: Storage-based Convergence of HPC and Cloud infrastructures to handle Big Data. Role: local coordinator for Inria Rennes Bretagne Atlantique.
Z-CloudFlow (2013-2016): geographically distributed workflows on Azure clouds. A project of the Microsoft-Inria Joint Research Centre.
Associated Teams
Associated Teams
SmartFastData (2019-2022): I was the French Principal Investigator of this joint team between Inria (KerData) and Instituto Politécnico Nacional, Mexico. The team focused on the data support for smart-cities. It financed several PhD internships and Invited Professors in both teams. Budget: 60K Euros.
Unify (2019-2022): I was involved in setting up and animating this joint team between Inria (KerData, DataMove) and Argonne National Laboratory (USA) focusing on flexible data support for scientific workflows.
Invited talks
Invited talks
Smart Big Data Seminar, Institituto Politécnico Nacional Ciudad de Mexico: The Fast Data as a means of HPC / Big Data / AI convergence, 2019
International Staff Week, Tbilissi State University: From Big Data to Fast Data, 2019
UPB Scientific Days, University Politehnica of Bucharest: The Path Towards the HPC / Big Data / IA convergence, 2019
UPB Scientific Days, University Politehnica of Bucharest: On the Challenges of the Edge/Cloud Hybrid Deployments, 2018
Huawei Workshop, Huawei Research, Munich: Low-Latency Stream Storage, 2017
BigStorage Summer School, FORTH Institute of Computer Science, Heraklion: Making Cities Smarter : A Storage-based View on Stream Processing Engines, 2017
UPB Scientific Days, University Politehnica of Bucharest: Science Driven, Scalable Data-Intensive Processing on Clouds, 2017
Smart Big Data Seminar, Institituto Politécnico Nacional Ciudad de Mexico: Science Driven, Scalable Data-Intensive Processing on Clouds, 2016
Distributed Systems Lab Seminar, University Politehnica of Bucarest: Big Data and Extreme Computing : A Storage-Based Pathway to Convergence, 2016
Future Cloud Symposium, EIT Digital Rennes: Enhancing video gaming user experience with Big Data analytics based on Apache Flink, 2015
Conf’Lunch, Inria / IRISA Rennes: Clouds and MapReduce Programming, 2015
IFB School on Cloud Computing, CumuloNumBio’15, Aussois: Big Data Management on Clouds, 2015
Distributed Systems Workshop, University Politehnica of Bucharest: Scalable Data Management on Clouds and Beyond, 2015
Trusted Cloud Workshop, EIT ICT Labs Rennes : Experience on running large scale scientific applications on public clouds, 2014
3rd Workshop on Storage and Cloud Computing, Technicolor Rennes: Scalable data management for scientific applications on cloud, 2013
Cloud Computing Seminar, EIT ICT Labs Rennes: Challenges of data storage on IaaS and PaaS clouds, 2013
High Performance Cloud and Big Data, Orange Labs Paris: On the issues of porting HPC applications to the clouds, 2013
Computing in the Cloud Workshop, EIT ICT Labs Rennes: Dealing with Big Data on clouds and beyond, 2013
IFIP Workshop, Futuroscope Poitiers : Cloud Computing : from technological advances to scientific challenges, 2013
Atelier France Grilles, Institut de Physique Nucléaire de Lyon: Gestion des données dans les clouds : l’approche BlobSeer, 2012
Séminaire BILab : Business Intelligence, Télécom-ParisTech: Stockage capable de passer à l’échelle pour les applications MapReduce, 2012
Séminaire Aristote : Big Data, École Polytechnique : Stockage à grande échelle pour les applications de traitement intensif des données, 2012
Page updated
Google Sites
Report abuse