HPE FlexFabric data center network plays a role in scientists’ quest to understand gravity

By Mary Gabra-Tanious, Blog Contributor
Share Post

The Max Planck Society, a Hewlett Packard Enterprise customer, is one of Germany's top research organization. More than 18 Nobel laureates have emerged from the ranks of its scientists, putting it on a par with the best and most prestigious research institutions worldwide.

Last February, the Max Planck Institute made big news with a stunning discovery that proves Einstein's 100-year-old theory about the existence of gravitational waves. BBC news summarized their findings in this article: Einstein's gravitational waves 'seen' from black holes.

Max Planck is a customer that I have been personally in touch with thru my role in promoting case studies for customers that use our networking equipment.  A year ago, I interviewed the team there to understand how they use our data center equipment and the benefits they realized.  With this news showing them as a Nobel Prize contending institute, I wanted to share with you how they are using the HPE data center networking solution.  To support  its research effort, the Institute built a massive computer cluster to gather and analyze data from the most sensitive gravitational wave detectors in the United States and Europe. Just to give you some figures on the magnitude of the data they work with, the cluster is made up of 3,350 compute nodes (most with 4 CPU cores), 37 file servers, and 12 storage servers, for an overall storage capacity of 4.4 petabytes.

Choosing Einsteinian simplicity and power

Max Planck's network was aging and due for a refresh. The team managing the infrastructure wanted to walk away from its existing proprietary environment to a more open network that would allow for flexibility and growth. Simplicity was key, but the project's leaders were also seeking a network architecture that would allow for high performance, reliability, and lower TCO – all of which HPE was able to offer with its data center networking solutions.

The Institute worked with HPE Partner microstaxx to design and build a 10Gbase-T network based on the HPE FlexFabric 12916 Switch AC Chassis for modular scalability and unprecedented levels of performance. "Einstein said things should be as simple as possible, but not simpler, and that's exactly what we've achieved with our new network," says Bruce Allen, Director of the Institute. "The reason we wanted the HP 12916 at our network core was the sheer power and simplicity of it."

Future science: opting for flexibility

Flexibility was another area that was crucial for Max Planck, to ensure that it can run its research in an efficient manner. "We have the largest capacity flagship core switch HPE sells, and with that comes all the flexibility we'll ever need," Allen says. "With a 16-slot chassis and 720 10Gb Ethernet ports, it's really a remarkable network core that will support whatever we want to do for the next 10 years."

Emphasizing how openness was a key need, Allen commented that "our HPE network is built on open standards for high reliability—this is telecommunications-grade stuff—and our experience shows that it's actually more reliable than our compute and storage environments."

 5 key business outcomes

With the HPE data center network, the Max Planck Institute is able to:

  • Maximize IT staff resources by simplifying network topology and maintenance
  • Enable seamless migration to new network fabric with zero unplanned downtime
  • Enhance its supercomputing environment with ease of adding new nodes, storage
  • Focus its innovation on research instead of IT
  • Meet tight timelines, delivering technology ahead of schedule and meeting academic budgets.

With Max Planck Institute making such remarkable discoveries, we at HPE are proud to be the providers of their data center network.

For the full Max Planck Institute case study, click here