Return to Workshops & Meetings

COLOC Open Workshop on Data Locality

 Data Locality, a critical aspect in modern heterogeneous systems


A well-known handicap for HPC applications running on modern highly parallelized and heterogeneous HPC platforms is that an increasing amount of time is spent in communication and data transfers; thus, the COLOC project, funded through the European ITEA-3 program, has been launched to design, implement and validate new approaches to optimize process placement and data locality management.
More precisely, the COLOC project focuses its efforts on (i) developing methods and tools enabling to model all resources of a computing platform using a hierarchical topology that describes the characteristics of these resources, (ii) enhancing upper software layers (resource manager such as SLURM; data communication libraries such as MPI; performance analysis tools such as MAQAO) to better manage data placement, thereby maximize application performance, and (iii) validating this approach, using applications from different domains: CFD (Computational Fluid Dynamics), CEM (Computational Electromagnetics), and CSM (Computational Structural Mechanics).

The aim of this session is to present the main technologies the project is advancing to improve the efficiency of modern HPC systems, so the proposed agenda is the following:

  • Short presentation of the COLOC project
  • HWLOC & TreeMatch: Cluster topology modeling for efficient exploitation of resources
  • MAQAO : Optimizing your application at node level
  • SLURM : Improving resources allocation to reduce data movement overhead in applications
  • Scotch : process placement and graph partitionning
  • Open Q/A session to get feedback

Keywords: Algorithm, Application, Parallel Programming, Performance, Scalability

Targeted audience:

  • HPC application developers interested in exploring new ways to optimize their code.
  • HPC centers and clusters managers to enhance cluster usage and application efficiency.
  • Academics and researchers in scientific computing.

Permanent link to this article: