NAStJA

Neoteric Autonomous Stencil code for Jolly Algorithms

NAStJA is a parallel framework for stencil code algorithms. NAStJA decomposes the simulated domain in small blocks and distributes those blocks over parallel working processing units. Interacting processing units build a local neighborhood and act autonomously in this neighborhood. This limits the number of connections for each processing unit and therefore the local communication overhead, and leads to high scalability.

Currently NAStJA is under development. An academic preview license is available.

Modules are available for the phase-field method, the phase-field crystal, and the cellular Potts model so far.

Publications about and with involvement of NAStJA

  • Non-Collective Scalable Global Network Based on Local Communications Berghoff, M.; Kondov, I. 2018. 9th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems (ScalA), Dallas, TX, USA, November 12, 2018, 25–32, IEEE, Piscataway (NJ)  doi:10.1109/ScalA.2018.00007
  • Massively Parallel Stencil Code Solver with Autonomous Adaptive Block Distribution Berghoff, M.; Kondov, I.; Hötzer, J. 2018. IEEE transactions on parallel and distributed systems.  doi:10.1109/TPDS.2018.2819672
  • Crystal-melt interface mobility in bcc Fe: Linking molecular dynamics to phase-field and phase-field crystal modeling Guerdane, M.; Berghoff, M. 2018. Physical review / B, 97 (14), 144105.  doi:10.1103/PhysRevB.97.144105
  • A Global Network for Non-Collective Communication in Autonomous Systems Berghoff, M.; Kondov, I. 2017. ICCS 2017, The International Conference on Computational Science, Zürich, Switzerland, 12th - 14th June 2017.  doi:10.5445/IR/1000079498

Visualization: The NAStJA Viewer

A small and fast open source visualization tool for 3D image data. Perfect for a fast look into simulation data. Just open the files and sweep around. A complicated setup of various filters is unnecessary. It supports NAStJA‘s output files and beyond other file formats it is easily extensible.

Features

  • Multi-platform support
  • VTI vtk image data (float32, float64, uint32, uint64)
  • CSV file for mapping origin data to several columns
  • Simple .dat files
  • Display of data in a block-structured grid
  • Time series
  • Different color bars
  • Discrete colors
  • 3D and 1D data view
  • Most options can be operated via keyboard