Media

Frank Leibiger

1 Comment

Supercomputing in the Open Telekom Cloud

  • Share
    Two clicks for more data privacy: click here to activate the button and send your recommendation. Data will be transfered as soon as the activation occurs.
  • Print
  • Read out
  • First public cloud in the world with integrated scalable supercomputing service
  • For maximum performance requirements and capability bursting in product development and research
170330-OpenTelekomCloud

Deutsche Telekom’s IT arm T-Systems has linked the Open Telekom Cloud with one of Europe´s top-end supercomputers. As a result, customers can integrate this high-performance computer at HLRS in Stuttgart in their scientific and engineering business workflows running in the Open Telekom Cloud.

“Our corporate customers only pay for the supercomputing resources they require. Something like this has not been seen on the market before,” explained Alfred Geiger, Managing Director at HWW*. “Other providers usually only provide dedicated supercomputing resources for longer binding periods.” HWW has been working for years at the interface between science and business. „For our clients it is important to know that we, based on our extensive experience, understand and support the security demands of enterprises,“ explained Bastian Koller, Managing Director of the supercomputing center.

The supercomputing center in Stuttgart is directly connected to the Open Telekom Cloud via a high-speed and secure line. As a result, users can effortlessly combine the public cloud and the supercomputing center.

More complex and faster

Such a combination covers the extensive computing requirements that arise, for example, in the field of computer-aided engineering. “At the moment, we are reaching our limits in conventional cloud environments such as the Open Telekom Cloud,” explains Geiger. The current limit is approximately 1,000 x86 cores. “At the supercomputing data center in Stuttgart, we can currently process workloads using over 180,000 cores, meaning this is up to 180 times more complex or faster.” This is due to the different architecture of the network. The cores must be able to communicate with each other with low latency and sufficient bandwidth at all times. Only in this way is the cluster able to work efficiently and to scale properly. However, the required packaging density and infrastructure goes beyond the scope of a “normal” cloud.

* Founded in 1995, HWW [supercomputing for science and industry] GmbH is a joint venture for the provisioning of HPC-resources of the High-Performance Computing Center (HLRS) in Stuttgart, T-Systems, Porsche AG, and the Karlsruhe Institute of Technology (KIT). HWW provides the commercial access to the compute-resources of HLRS and KIT on a pay-per-use base for European academic institutions and industrial companies, usually for industrial research and development purposes.

About Deutsche Telekom: Company profile
About T-Systems: T-Systems company profile

 It's happening in the cloud.

Cloud Computing

It's happening in the cloud.

FAQ