Section: New Software and Platforms
Damaris
- Title:
-
Damaris: I/O and data management for large-scale, MPI-based HPC simulations.
- Keywords:
-
I/O; HPC; Data management; Visualization; Big Data; Exascale.
- Scientific Description:
-
Damaris is a middleware for multicore SMP nodes enabling them to efficiently handle data transfers for storage and visualization. The key idea is to dedicate one or a few cores of each SMP node to the application I/O. It is developed within the framework of a collaboration between KerData and the JLESC. The current version enables efficient asynchronous I/O, hiding all I/O-related overheads such as data compression and post-processing, as well as direct (in situ) interactive visualization of the generated data.
Damaris has been preliminarily evaluated at NCSA (Urbana-Champaign) with the CM1 tornado simulation code. CM1 is one of the target applications of the Blue Waters supercomputer in production at NCSA/UIUC (USA), in the framework of the JLESC. Damaris now has external users, including (to our knowledge) visualization specialists from NCSA, Big Data experts from the HDF group, and researchers from the France/Brazil Associated Research Team on Parallel Computing (joint team between Inria/LIG Grenoble and the UFRGS in Brazil). Damaris has been successfully integrated into four large-scale simulations (CM1, OLAM, Nek5000, CROCO). Works are in progress to evaluate it in the context of several other simulation codes including HACC (cosmology) and GTC (fusion).
Damaris is the object of a Technical Development Action (ADT) supported by Inria.
- Functional Description:
-
Damaris targets large-scale HPC simulations: in situ data analysis by some dedicated cores of the simulation platform; asynchronous and fast data transfer from HPC simulations to Damaris; semantic-aware dataset processing through Damaris plug-ins.
- Contact data:
-
- Participants:
-
Matthieu Dorier (ANL), Lokman Rahmani, Gabriel Antoniu, Orçun Yildiz, Hadi Salimi and Luc Bougé.
- Partners:
- Contact:
- URL: