Ralf Kundel, M.Sc.

Group Head


Phone: +49 (0) 6151 16 20855
E-Mail: Ralf.Kundel@...
Location: S3|20 216

ORCiD: 0000-0003-1711-5990

8 Private Questions (in german)

 

 

Short Bio

Ralf obtained his B.Sc. and M.Sc. degree from Technical University of Darmstadt, Germany in 2015 and 2017 respectively. Since 2018 he is a Ph.D. student at the multimedia communications lab.
In his research he focuses on quality of service aware network functions offloaded on programmable hardware such as P4-programmable switches and FPGAs. One main focus is on internet service provider access networks. He works together with Deutsche Telekom in joined projects opening up network hardware for improving the future Internet.

Research Interests

  • Hardware Acceleration
  • FPGAs
  • P4
  • Network Optimization
  • Low Latency Networking
  • Active Queue Management (AQM)
  • ISP access networks (residential and 5G)

 

Involved Projects:

  • Deutsche Telekom: Dynamic Networks. This project focus on applying concepts of SDN and programmable hardware in carier grade networks. Specially technologies such as FPGAs and P4 programmable switches but also concepts like Network Functions Virtualization (NFV) allow novel approaches of rethinking access networks.
  • BMBF: Software Campus 5G-PCI. Hardware acceleration is becoming very important in order to meet high performance and low energy consumption at the same time. This project investagates how PCIe accelerators can be virtualized and shared between multible virtual borrowers.
  • BMBF: Open 6G Hub. Providing a high performant communication infrastructure has become of tremendous importance. As part of this hub, we investigate new concepts for providing more resilient 5G/6G networks. One key component of our investigation is the disaggregated O-RAN architecture.
  • BMBF: AI-NET-PROTECT.  In this project, novel approaches for resilient and secure network operation are investigated. The last decade has shown that providing only performant networks is not sufficient to ensure a reliable communication infrastructure. Building upon machine learning approaches, this project aims to enhance the security of future networks.
  • DFG: Collaborative Research Centre 1053 MAKI (associated researcher). The Internet has grown to a gigantic computer network which includes many protocols and technologies. This research centre investigates how multiple mechanisms (i.e. protocols and technologies) can be used best together and how a transition between those can be performed at the best possible performance.

 

Open Source Contributions:

  • P4-CoDel  Implementation of CoDel AQM for Barefoot Tofino, Netronome SmartNIC and bmv2 software switch.
  • P4STA Load generation and aggregation framework based on programmable networking hardware.
  • free5gc fork of existing project with improvements to work with real 5G standalone RAN networks and UEs.

 

 

Publications