Teaching

Teaching

My main teaching are System and Large Scale Distributed Systems Some material@Toulouse: M1 parallelism Introduction lesson MPI Lesson MPI and MPI Python examples from the Lesson Example of collective operations TD MPI Labs MPI, Files for the labs Homework MPI, Files for the Homework Monitors Lesson 1, Lesson 2 Monitor Exercises: TD1 TD2 TD3 Monitor Labs and Python sheet for labs M1 Systèmes répartis Project: Multi-player fully distributed game lab M2 resource management

Mobile Wireless Network@Addis Abeba

Mobile Wireless Network teached at Addis Abeba University Teaching at doctoral school of Addis Abeba Lessons Cours 1: Mobile Wireless Networks Practical session 1: Low level network tools (with ethereal.cap and ethereal_traceroute.cap) Cours 2: Peer to peer 1 Practical session 2: Gnutella Cours 3: Peer to peer 2 Practical session 3: Chord Articles for the presentation session Adaptive Clustering for Mobile Wireless Networks Broadcasting in Vanet GloMoSim A Library for Parallel Simulation of Large-scale Wireless Networks Maximum Battery Life Routing to Support Ubiquitous Mobile Computing in Wireless Ad Hoc Networks MLSR A Novel Routing Algorithm for Multilayered Satellite IP Networks NCTUns network simulation and emulation for wireless resource management Protocols for Adaptive Wireless and Mobile Networking SEAD secure efficient distance vector routing for mobile wireless ad hoc networks SWAN A Mobile Multimedia Wireless Network

MPI Collective operations

bcast Play/Replay bcast scatter Play/Replay scatter gather Play/Replay gather reduce Play/Replay reduce all_gather Play/Replay all_gather all_reduce Play/Replay all_reduce

Multi-player fully distributed game lab

Distributed systems The goal of these labs is to implement several synchronization algorithms. For this we will use a small multi-player board game. The provided version does not manage the synchronization between players. It leads to a final state where several players have a different view on the results. You will have to implement several algorithms seen during the lessons. Centralized approach (as baseline) Token-based approach Ricart and Agrawala mutual exclusion Ernest Chang and Rosemary Roberts.

Sujet énergie pour Systèmes Embarqués

Système Embarqué - Énergie Accès à Grid5000 Slides d’introduction Getting Started où on vous explique comment accéder à distance en rajoutant le texte suivant dans votre fichier .ssh/config Host g5k User login Hostname access.grid5000.fr ForwardAgent no Host *.g5k User login ProxyCommand ssh g5k -W "$(basename %h .g5k):%p" ForwardAgent no Commande pour sélectionner un noeud sur le cluster Gros à Nancy ssh nancy.g5k oarsub -I -l walltime=2:0:0 -p "cluster='gros'" Séance 1 et 2 : Slides et prise en main MojitO/S Installer MojitO/S (monitoring système) : https://gitlab.