No abstract

General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/userguides/explore-bristol-research/ebr-terms/ 1932-4537 (c)Abstract-Datacenter (DC) design has been moved towards the edge computing paradigm motivated by the need of bringing cloud resources closer to end users. However, the Software Defined Networking (SDN) architecture offers no clue to the design of Micro Datacenters (MDC) for meeting complex and stringent requirements from next generation 5G networks. This is because canonical SDN lacks a clear distinction between functional network parts, such as core and edge elements. Besides, there is no decoupling between the routing and the network policy. In this paper, we introduce Residue Defined Networking Architecture (RDNA) as a new approach for enabling key features like ultra-reliable and low-latency communication in MDC networks. RDNA explores the programmability of Residues Number System (RNS) as a fundamental concept to define a minimalist forwarding model for core nodes. Instead of forwarding packets based on classical table lookup operations, core nodes are tableless switches that forward packets using merely remainder of the division (modulo) operations. By solving a residue congruence system representing a network topology, we found out the algorithms and their mathematical properties to design RDNA's routing system that (i) supports unicast and multicast communication, (ii) provides resilient routes with protection for the entire route, and (iii) is scalable for 2-tier Clos topologies. Experimental implementations on Mininet and NetFPGA SUME show that RDNA achieves 600 ns switching latency per hop with virtually no jitter at core nodes and submillisecond failure recovery time.

The increased carrier bandwidth and the number of antenna elements expected in 5G networks require a redesign of the traditional IP-based backhaul and CPRI-based fronthaul interfaces used in 4G networks. We envision future mobile networks to encompass these legacy interfaces together with novel 5G RAN functional splits. In this scenario, a consistent transport network architecture able to jointly support backhaul and 4G/5G fronthaul interfaces is of paramount importance. In this paper we present 5G-XHaul, a novel transport network architecture featuring wireless and optical technologies and a multi-technology software defined control plane, which is able to jointly support backhaul and fronthaul services. We have deployed and validated the 5G-XHaul architecture in a city-wide testbed in Bristol.

The ability of scaling power and performance at run-time enables the creation of computing systems in which energy is consumed in proportion of the work to be done and the time available to do it. These systems favour active energy-efficient states in which useful computation is performed at low energy instead of using inactive energy savings modes that incur large latency and energy penalties to enter and exit modes in which the system is halted. This is particular useful in servers that spend most of their time at around 30% utilization and are rarely fully idle or at maximum utilization. A feature of an energy proportional computing system is that it must exhibit a wide dynamic range with multiple levels of energy and performance available. In this context this paper investigates how these levels can be obtained in commercially available state-of-the-art 28 nm FPGAs and characterizes its benefits. Adaptive voltage and frequency scaling is employed to deliver proportional performance and power in these FPGA devices. The results reveal that the available voltage and frequency margins create a large number of performance and energy states with scaling possible at run-time with low overheads. Power savings of up to 64.98% are possible maintaining the original performance at a lower voltage.

No abstract

Virtual Data Centre (VDC) solutions provide an environment that is able to quickly scale-up and where virtual machines and network resources can be quickly added ondemand through self-service procedures. VDC providers must support multiple simultaneous tenants with isolated networks on the same physical substrate. The provider must make efficient use of their available physical resources whilst providing high bandwidth and low-latency connections to tenants with a variety of VDC configurations. This paper utilises state of the art optical network elements to provide high bandwidth optical interconnections and develop an VDC architecture to slice the network and the compute resources dynamically, to efficiently divide the physical network between tenants. We present a Data Centre Virtualisation architecture with an SDN-controlled all-optical data plane combining Optical Circuit Switching (OCS) and Time Shared Optical Network (TSON). Developed network orchestration dynamically translates and provisions VDCs requests onto the optical physical layer. The experimental results show the provisioned bandwidth can be varied by adjusting the number of time slots allocated in the TDM network. These results lead to recommendations for provisioning TDM connections with different performance characteristics. Moreover, application level optical switch reconfiguration time is also evaluated to fully understand the impact on application performance in VDC provision. The experimental demonstration confirmed the developed VDC approach introduces negligible delay and complexity on the network side.

This paper reports an FPGA-based P4-enabled Smart NIC solution which is designed and implemented for webscale cloud and to meet 5G/Beyond 5G networking requirements. The P4-enabled Smart NIC solution leverages the open standards, platforms and software-defined approaches, responds to the real time Data Centre Networking service requests, in particularly, enables the end-to-end network slicing, which is one of the critical requirements of multi-tenancy 5G network. We discussed the possibilities and challenges of P4 specification implementation in the FPGA to realise the Smart NIC functionalities. And after that, we showed its data plane programmability and flexibility with P4 features. Furthermore, we demonstrated its application scenario in an 5G environment mainly focusing on edge Data Centre to core Data Centre network slicing. The setup interconnects the P4enabled Smart NIC with optical Bandwidth Variable Transponders, and the system offers agile 100Gbps interface to transport the packets through P4-defined data plane for L2/L3/L4 parsing and action. The P4-enabled Smart NIC can change the data plane pipelines in seconds, and it can achieve maximum 84.8Gbps throughput. With P4 programmed hardware offloaded Segment Routing can produce 30% more bandwidth than without.

We present a Data Centre Virtualisation architecture with an SDN-controlled all-optical data plane combining OCS and TSON. Orchestration dynamically translates and provisions Virtual Data Centres requests onto the optical layer. We describe an implementation and characterisation of the data plane.