HeteroPar'2016 Program

09:20-09:30 Opening words - Loris Marchal
09:30-10:30 Georges Da Costa, Université Toulouse III. Harnessing heterogenous architectures for green cloud datacenters (slides)
Session 1
11:00-11:30 Terry Cojean, Abdou Guermouche, Andra-Ecaterina Hugo, Raymond Namyst and Pierre-André Wacrenier. Resource aggregation in task-based applications over accelerator-based multicore machines
11:30-12:00 Joăo F. D. Guerreiro, Aleksandar Ilic, Nuno Roma and Pedro Tomás. Performance and Power-Aware Classification for Frequency Scaling of GPGPU Applications.
12:00-12:30 Emmanuel Agullo, Luc Giraud, Abdou Guermouche, Stojce Nakov and Jean Roman. Task-based Conjugate Gradient: from multi-GPU towards heterogeneous architectures
Session 2
14:30-15:00 Emmanuel Agullo, Luc Giraud and Stojce Nakov. Task-based sparse hybrid linear solver for distributed memory heterogeneous architectures
15:00-15:30 Shin Morishima and Hiroki Matsutani. Distributed In-GPU Data Cache for Document-Oriented Data Store via PCIe over 10Gbit Ethernet
15:30-16:00 José Ignacio Aliaga, Ernesto Dufrechou, Pablo Ezzatti and Enrique S. Quintana-Orti. A Data-Parallel ILUPACK for Sparse General and Symmetric Indefinite Linear Systems
Session 3
16:30-17:00 Sergio Afonso, Alejandro Acosta and Francisco Almeida. Automatic Generation of OpenCL Code for ARM Architectures
17:00-17:30 Dariusz Król, Rafael Ferreira Da Silva, Ewa Deelman and Vickie Lynch. Workflow Performance Profiles: Development and Analysis