Ahcene Bendjoudi
Ing, PhD,

Full-time researcher at CERIST Research Center, Algiers, Algeria.

Main research interests

My research interests focus on:
    • High Performance Computing & Parallel Optimization
        • GPU Computing/Multi-core programming
        • Grid/Cluster and P2P - Computing
        • Parallel algorithms for Combinatorial Optimization
    • Big Graphs
        • Framworks and HPC Platforms for Big Graphs
        • Big Graphs Partitioning
        • Big Graphs Querying
    • Smart Cities
        • Dynamic Route Planning
        • Smart Mobility
        • Disaster Management

Research Projects

    • Leader of the project (HPC-Optim) High Performance Computing for Combinatorial Optimization, CERIST, March 2012-December 2016.
    • Member (WP Leader) of the project (BiGraPS) Big Graph Data Processing on HPC platforms: Application to improve Mobility in Smart Cities, January 2017-December 2019.
    • Member of the coordinating team of ERANETMED, December 2013 - December 2017.

Collaboration

    • MCSI, ESI, Algiers, Algeria
        • Hybrid GPU/CPU multi-core B&B algorithms
        • Association Rule Mining
        • Parallel satellite image processing
    • DOLPHIN Team, INRIA/LIFL/University Lille1, Lille, France.
        • Grid-based Parallel Branch and Bound Algorithms
        • GPU/multi-core B&B algorithms

International Responsabilities

Reviewer in International Journals

    • IEEE TPDS - IEEE’s Transactions on Parallel and Distributed Systems
    • JPDC - The Journal of Parallel and Distributed Computing
    • CCPE - Concurrency and Computation: Practice and Experience
    • PPL - Parallel Processing Letters
    • SOCO - Soft Computing Journal
    • CAIE - Computers & Industrial Engineering
    • IJICT - International Journal of Information and Communication Technology

Conference/Event Organization

    • Member of the steering committee of IEEE ICT-DM'2015 ICT-DM'2015, France, December 2015.
    • Organizer of HPC days (JCIA'2014) Algiers, October 2014.
    • Member of the organizing committee of ICT-DM'2014, Algeria, March 2014.
    • Chair of the Scientific Computing training session and member of organizing committee of CARI’2012 (Algiers 2012)

Member of Technical Program Committees

    • IEEE Int. Conf. on ICTs for Disaster Management, ICT-DM'2017, Germany, December 2017.
    • Workshop on Parallel Optimization using/for Multi and Manycore HPC (POMCO'2016), Austria, 2016
    • 13th Int. Conf. on Distributed Computing and Artificial Intelligence (DCAI'16), Spain, 2016
    • IEEE Int. Conf. on Vehicular Traffic Management (IEEE VTM’2016)
    • Multi/Manycore computing for parallel Metaheuristics (McM’2015), Morocco, 2015
    • International Workshop on Parallel Optimization using High Performance Computing, POMM'2014, Bologna, Italy, 2014
    • International Conference on Metaheuristics and Nature Inspired Computing, META'2014, Marrakech, Morocco, 2014
    • International Conference on Advanced Networking, Distributed Systems and Applications INDS'2014, Bejaia, Algeria, 2014
    • IEEE International Conference on Vehicular Traffic Management IEEE VTM'2014, Seoul, Korea, 2014
    • IEEE International Conference on Information and Communication Technologies for Disaster Management IEEE ICT- DM'2014, Algiers, Algeria, 2014
    • IEEE International Conference on Vehicular Traffic Management IEEE VTM'2012, Dublin, Ireland, 2012

Thesis summary

Solving to optimality large instances of combinatorial optimization problems using Branch and Bound (B&B) algorithms requires a huge amount of computing resources. Nowadays, such power is provided by large scale environments such as computational grids. However, grids induce new challenges: scalability, heterogeneity, and fault tolerance. Most of existing grid-based B&Bs are developed using the Master-Worker paradigm, their scalability is therefore limited. Moreover fault tolerance is rarely addressed in these works. In this thesis, we propose three main contributions to deal with these issues: P2P-B&B, H-B&B, and FTH-B&B. P2PB&B is a MW-based B&B framework which deals with scalability by reducing the task request frequency and enabling direct communication between workers. H-B&B also deals with scalability. Unlike the state-of-the-art approaches, H-B&B is fully dynamic and adaptive, meaning it takes into account the dynamic acquisition of new computing resources. FTH-B&B is based on new fault tolerant mechanisms enabling efficient building of the hierarchy and maintaining its balancing, and minimizing of work redundancy when storing and recovering tasks. The proposed approaches have been implemented using ProActive grid-middleware and applied to the Flow-Shop scheduling Problem (FSP). The large scale experiments performed on Grid'5000 proved the efficiency of the proposed approaches.

Ph.D. dissertation