ICSA Colloquium - 18/11/2021

Title: Teaching the 5G vRAN to share compute with Concordia


Abstract: Virtualized Radio Access Networks (vRAN) offer a cost-efficient solution for running the 5G RAN as a virtualized network function (VNF) on commodity hardware. The vRAN is more efficient than traditional RANs, as it multiplexes several base station workloads on the same compute hardware. While this multiplexing provides efficiency gains, more than 50% of the CPU cycles in typical vRAN settings still remain unused. A way to further improve CPU utilization is to collocate the vRAN with general-purpose workloads. However, to maintain performance, vRAN tasks have sub-millisecond latency requirements that have to be met 99.999% of times. In this talk, I will demonstrate why this is difficult to achieve with existing systems through experimental results obtained using a commercial-grade reference vRAN solution. I will then present Concordia, a userspace deadline scheduling framework for the vRAN that is specifically designed to enable the collocation of other best-effort workloads with the vRAN on commodity Linux-based platforms. Concordia builds prediction models using quantile decision trees to predict the worst case execution times of vRAN signal processing tasks. The Concordia scheduler is fast (runs every 20 microseconds) and the prediction models are accurate, enabling the system to reserve a minimum number of cores required for vRAN tasks, leaving the rest for general-purpose workloads. I will conclude the talk by demonstrating the benefits of Concordia in meeting the 99.999% reliability requirements, while reclaiming up to 80% of idle CPU cycles without affecting the RAN performance on a commercial-grade reference vRAN platform.


Bio: Xenofon Foukas is a Senior Researcher at Microsoft Azure for Operators and is a member of the Office of the CTO team. Prior to this role, he was a Senior Researcher at Microsoft Research Cambridge and a Research Associate in the School of Informatics at the University of Edinburgh. He received his PhD in 2018 from the University of Edinburgh and his MSc from Imperial College London. His research interests are in the intersection of 5G and beyond mobile network architectures, edge computing and applied machine learning for mobile networks. His current research focuses on designing efficient, scalable and intelligent virtualized Radio Access Networks.

Nov 18 2021 -

ICSA Colloquium - 18/11/2021

Xenofon Foukas - Microsoft

Hybrid (G.03, IF or Zoom)