Curriculum Vitae


Mr. Everton HERMANN

 

52  Rue d'Hautpoul

75019, Paris

France

Nationality : Brazilian

Date of Birth : 04/08/1982

E-mail : everton.hermann@gmail.com

Web: http://evertonhermann.tk

R&D

Engineer

High Performance Computing

Areas of Interest

High Performance Computing

Operating Systems

Distributed Systems

Networks and Web

Physics Simulation

Education

2006-2010 : PhD in Computer Science - Institut National Polytechnique de Grenoble

Subject: Interactive Physical Simulation on Multi-core and Multi-GPU Architectures.

Multi grain parallelization, Multi-GPU, Multi-CPU, Load Balancing

2005-2006 : Master in Computer Science  – Universidade Federal do Rio Grande do Sul

Subject:  Data Server  Dynamism on the dNFSp File System  (NFS, Lustre, PVFS)

Parallel and distributed file system, high performance communication

                    

2000-2004 : BSc. in Computer Science  - Universidade Federal de Santa Maria   

Final Semester Project:  SFTPFS,  A SSH/SFTP File System for the  GNU HURD Operating System.

                     

2003-2004 : Exchange Student – 3rd Year in Computer Science, ENSIMAG, INPG.   

Specialization: Distributed Systems and Networks.

Final Semester Project: Porting the INUKTITUT  library to Myrinet  network and Itanium-2 

 High performance networks, cluster computing

Work and Teaching Experiences

2010:  R&D Engineer at Ubisoft Paris

High Performance Computing

Optimization of a post production pipeline for computer generated movies. (C++, OpenCL)

2009: Teaching Assistant at ENSIMAG:

Second year students : Concurrent programming and Operating Systems ( pthreads, MPI)

First year students: Networks

Third year students: Network Project

2006 – 2008 : Teaching Assistant at ENSIMAG:

Second year students : Concurrent programming and Operating Systems ( pthreads, MPI)

2005 – 2006 :  Teaching Assistant at UFRGS:

Undergraduate students : Fundamentals of Fault Tolerance(Fault Tolerant Storage and Computing)

Internships

08/2004 – 03/2005 :    Hewlett-Packard /GPPD, Universidade Federal do Rio Grande do Sul (UFRGS)

- Development of the dNFSP  distributed file system (C)

- Testing and performance evaluation

- Cluster Management

02/2004 – 07/2004 :    INRIA, Laboratoire Informatique et Distribution

- Development of parallel and distributed computing tools (C++)

- High Performance Network Communication (C, Myrinet GM)

Competencies

Language Skills:

 Fluent: English, French, Spanish, Portuguese (mother tongue)

 Basic knowledge: German, Italian

Programming and Scripting Languages :

Awk, C, C++, CUDA, Java,  Pascal, PHP, Perl, Shell Script,  Python, SQL

Operating Systems :

AIX, xBSD, GNU HURD, GNU/Linux, IBM zVM, Windows

International Conferences Publications

2010:         E. Hermann, B. Raffin, F. Faure, T. Gautier and J. Allard:

                  Multi-GPU and Multi-CPU Parallelization for Interactive Physics Simulations.

                  Europar 2010

2009 :         E. Hermann, B. Raffin, F. Faure.

                  Interactive Physical Simulation on Multicore Architectures.

                  9th Eurographics Symposium on Parallel Graphics and Visualization.

2008 :         E. Hermann, B. Raffin, F. Faure.

                  Ray-traced collision detection for deformable bodies.

                  3rd  International Conference on Computer Graphics Theory and Applications.

2007 :         E. Hermann, R. Kassick, R. Avila, C. Barrios and M. Riveill , Y. Denneulin,P. Navaux.

                  Performance Evaluation of Meta-Data Transfer and Storage in Clusters.

                      Latinamerican Conference of High Performance Computing.

2006 :          E. Hermann, R. Ávila, P. Navaux, and Y. Denneulin.

                      Metaserver Locality and Scalability in a Distributed NFS.

                      In 6th International Meeting on High Performance Computing for    Computational Science.

Projects

SOFA Framework  :  Open Source framework  targeted at real-time simulation (C++, CUDA)

KAAPI                       :  Middleware for adaptive computation on multi-core architectures clusters and grids (C++)

NFSp                              :  A Non-Intrusive Parallel NFS Server (C)

Personal Interests

Traveling, Languages, Music (Flute, Guitar, Bass, Piano),  Brewing, Soccer, Hiking, Biking