1 / 2

Scalable Performance Analysis Toolset for High-Performance Computing (HPC)

The Scalasca performance analysis toolset is designed for parallel codes and is compatible with a wide range of High-Performance Computing (HPC) platforms. It offers various analysis modes, providing performance overviews and in-depth studies of communication and synchronization through automatic trace analysis. Key features include wait state identification and support for programming models such as MPI 2.2 and OpenMP. Scalasca is available under the New BSD license and can be installed at organizations like LLNL for testing purposes.

tait
Télécharger la présentation

Scalable Performance Analysis Toolset for High-Performance Computing (HPC)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Scalable performance-analysis toolsetfor parallel codes • Suitable for wide range of HPC platforms • Analysis modes • Performance overview on call-path level via runtime summarization • In-depth study of communication & synchronization via automatic trace analysis • Identification of wait states • Supported programming models • MPI 2.2 including RMA analysis • OpenMP (basic features) • Available under the New BSD license • http://www.scalasca.org/ • Installed at LLNL for TLCC, up, and BG/L(unsupported test version) Joint project of

  2. Libra: Scalable Load Balance Analysis Running Application • Libra is a tool for quickly monitoring load-balance • Handles applications with large numbers of time steps • Handles time-evolving patterns • Small, phased effort profiles displayed intuitively in surface plot • Can observe different phases of code separately • Scalable wavelet-based compression enables use at scale EffortInstrumentation ScalableCompression ScalableVisualization

More Related