• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Artikel
  4. D3: Enhancing reconfigurable datacenters with adaptive demand-oblivious and demand-aware integration
 
  • Details
  • Full
Options
June 2026
Journal Article
Title

D3: Enhancing reconfigurable datacenters with adaptive demand-oblivious and demand-aware integration

Abstract
The explosively growing communication traffic in datacenters imposes stringent performance requirements on the underlying networks. Over the last years, researchers have developed innovative optical switching technologies that enable reconfigurable datacenter networks (RDCNs) with fast topology reconfigurations. This paper presents D3, a novel RDCN architecture suited for PoD-level deployment, improving throughput and flow completion time and enabling scalable composition within hyperscale datacenter networks. D3 adapts its links and packet scheduling toward the evolving demand, combining both demand-oblivious and demand-aware behaviors. D3 relies on a decentralized network control-plane supporting greedy, integrated-multihop, IP-based routing, allowing to react quickly and locally to topological changes without overheads. Moreover, we argue that D3 can be implemented using recently proposed optical-switching architectures. Using extensive evaluations, we find that D3 improves throughput by up to 15% and preserves competitive flow completion times compared to the state-of-the-art. We further provide an analytical explanation of the superiority of D3, introducing an extension of the well-known Birkhoff-von Neumann decomposition, which may be of independent interest.
Author(s)
Zerwas, Johannes
Griner, Chen
Schmid, Stefan  
Fraunhofer-Institut für Sichere Informationstechnologie SIT  
Avin, Chen
Journal
Computer Networks  
DOI
10.1016/j.comnet.2026.112255
Language
English
Fraunhofer-Institut für Sichere Informationstechnologie SIT  
  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024