• Project Overview
  • Consortium
  • Advisory Board
  • News & events
    • News
    • Events
  • Publications & Media
    • Gallery
    • Publications
    • Public Results
    • Newsletters & Press Releases
    • Videos
  • Repositories & Links
    • Related Projects
    • Cloud benchmarks
    • Related Organisations
    • VINEYARD's Repositories
  • Contact
  • Project Overview
  • Consortium
  • Advisory Board
  • News & events
  • Publications & Media
  • Repositories & Links
  • Contact
See all news

VINEYARD leveraging from Maxeler’s new MAX5 dataflow engines


VINEYARD will be taking advantage from Maxeler's fifth generation DFE, MAX5 system, recently installed at VINEYARD’s coordinator premises (ICCS), a further system will be also available to STFC soon. These systems will be used in the project to develop dataflow accelerated applications and as a testbed in the final evaluation.

The new DFE MAX 5 uses the powerful VU9P compute chip from Xilinx and sets a new standard in multiscale dataflow computing. MAX5 brings a 3-5x performance improvement and 2-3x increase in energy efficiency over previous generation MAX4 DFEs. Several leading customers have already upgraded their existing systems and are benefiting from its enhanced capabilities. MAX5 DFEs have been deployed in an HPC system at the Jülich Supercomputer Centre and are also being adopted by an increasing number of academic partners.

Last but not least, MAX5 DFEs are fully compatible with the Amazon EC2 F1 instances from Amazon Web Services. Maxeler supports the development and migration of applications to Amazon F1 with its MaxCompiler software.

Share this

Share on LinkedIn Share on Twitter

  • Gallery

    VINEYARD project in images

  • Publications & Media

    Get the official VINEYARD documents

  • Repositories & Links

    These are some useful links to show you more about VINEYARD’s framework

  • Contact

    For more information, contact us

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 687628

Subscribe Newsletter
Share on LinkedIn Share on Twitter
LOBA