Packet processing on Linux can be slow due to its complex network stack. To solve this problem, there are two main solutions: eXpress Data Path (XDP) and Data Plane Development Kit (DPDK). XDP and the AF _XDP socket offer full interoperability with the legacy system and is being adopted by major internet players like Open vSwitch or Facebook. While the performance evaluation of AF _XDP against the legacy protocol stack in the kernel or against DPDK has been studied in the literature, the impact of the multiple socket parameters and the system configuration on its latency has been left aside. To address this, we conduct an experimental study to understand the XDP/AF _XDP ecosystem and detect microseconds delays to better architect future latency-sensitive applications. Since the performance of AF _XDP depends on multiple parameters found in different layers, finding the configuration minimizing its latency is a challenging task. We rely on a classification algorithm to group the performance results, allowing us to easily identify parameters with the biggest impact on performance at different loads. Last, but not least, we show that some configurations can significantly decrease the benefits of AF _XDP, leading to undesirable behaviors, while other configurations are able to reduce such round trip delays to an impressive value of 6.5 μS in the best case, including the tracing overhead. In summary, AF _XDP is a promising solution, and careful selection of both application and socket parameters can significantly improve performance.

Maintaining up-to-date KVM hypervisors requires regular upgrades to the host kernel, hence rebooting the physical host with the consequent termination of running Virtual Machines (VMs). Cloud platforms capable of massive large-scale live migrations evacuate VMs from the hosts before rebooting, minimizing the impact over VM up-time. However, scenarios exist where resource constraints make live migration undesirable, or the presence of fault-tolerant instances (e.g., replicated services) favors the adoption of VM termination, a simpler but more disruptive strategy. In this article, we present Hy-FiX, a fast in-place upgrade solution for KVM hypervisors. Hy-FiX preserves VM memory across host reboots, protecting the execution state of running guests while hypervisor upgrades are applied. Hy-FiX memory preservation across reboot, combined with a mixed suspend-to-disk/suspend-to-RAM technique, achieves a 2.31-second checkpoint/restore time for a 256 GB VM, and Hy-FiX lazy memory initialization reboots an enterprise-class host in constant time (7.6 seconds) regardless of its equipped memory. Hy-FiX is, therefore, a better alternative to classical VM termination and restart.

Live video distribution represents a significant fraction of the video traffic in the Internet. Its distribution induces heavy costs to serve clients during peak hours. To tackle this issue, the hybrid Viewer-to-Viewer (V2V) mode - where peers cooperate with each other in addition to receiving data from a Content Distribution Network (CDN) - has been proposed to both improve the Quality of Experience (QoE) and reduce the distribution cost. In this paper, we present the improvements we have introduced in a commercial V2V Web live streaming system based on the Web Real-Time Communication (WebRTC) standard. A key challenge faced when designing such a system is the algorithm that governs the client's behavior and should maximize the amount of a data exchanged in V2V mode. A crucial component in V2V Web live streaming is the fast distribution of the video chunks availability in the V2V network. We present in this work our initial reactive approach to distribute the information concerning chunks availability and contrast it with our new proactive approach. We demonstrate the efficiency of the proactive approach on a testbed featuring light clients (browsers) separated using Linux network namespaces, as well as with live experiments. The fraction of V2V traffic for the production environment also improved from 19% in the reactive approach to 37% in the proactive one. Furthermore, we demonstrate that the buffer length of the clients both in the controlled environment and in production is improved, implying a better QoE.

Hypervisors are critical components in cloud data centers. Maintaining up-to-date KVM hypervisors requires replacing the host kernel, hence rebooting the physical host with consequent termination of running virtual machines (VMs). We introduce Hy-FiX, an in-place upgrade tool for KVM hypervisors that applies host kernel and/or user-space components upgrades while transparently preserving running VMs, constituting a better alternative for platforms where VMs are inevitably terminated and restarted.

In Data Centers (DCs), an abundance of virtual machines (VMs) remain idle due to network services awaiting for incoming connections, or due to established-and-idling sessions. These VMs lead to wastage of RAM - the scarcest resource in DCs - as they lock their allocated memory. In this paper, we introduce SEaMLESS, a solution designed to (i) transform fully-fledged idle VMs into lightweight and resourceless virtual network functions (VNFs), then (ii) reduces the allocated memory to those idle VMs. By replacing idle VMs with VNFs, SEaMLESS provides fast VM restoration upon user activity detection, thereby introducing limited impact on the Quality of Experience (QoE). Our results show that SEaMLESS can consolidate hundreds of VMs as VNFs onto one single machine. SEaMLESS is thus able to release the majority of the memory allocated to idle VMs. This freed memory can then be reassigned to new VMs, or lead to massive consolidation, to enable a better utilization of DC resources.

Idle virtual machines (VMs) are a waste of resources in data centers. We introduce SEaMLESS, which transforms a fully-Hedged idle VM into a lightweight and resourceless Virtual Network Function (VNF). Idle VMs can then be saved to disk and release their memory. Simultaneously, the VNF provides service availability. Upon user activity, the appropriate VM is restored, without introducing any interruption for service users. Tens of VNFs can be contained within the same memory space required for one single VM, thereby facilitating ample resources savings when scaled up to a data center.

Networking infrastructures are considered to consume as much energy as terminal end-user equipment or datacenters. While energy consumption of wireless networks is a matter of concern since their beginning, it is not the case for wired networks as they do not rely on batteries, but on plugged equipment. Yet, facing growing consumption, energy-efficient techniques start to be implemented in wired networks. However, measuring the end-to-end energy consumption of wired networking infrastructures remains a real challenge for network operators and scientists. This article presents the ECOFEN (Energy Consumption mOdel For End-to-end Networks) framework which allows to support precise simulation of energy consumption of large-scale complex wired networks. The experimental validation shows that Ecofen provides accurate energy consumption values.

Energy aware routing aims at reducing the energy consumption of ISP networks. The idea is to adapt routing to the traffic load in order to turn off some hardware. However, it implies to make dynamic changes to routing configurations which is almost impossible with legacy protocols. The Software Defined Network (SDN) paradigm bears the promise of allowing a dynamic optimization with its centralized controller. In this work, we propose SENAtoR, an algorithm to enable energy aware routing in a scenario of progressive migration from legacy to SDN hardware. Since in real life, turning off network equipments is a delicate task as it can lead to packet losses, SENAtoR provides also several features to safely enable energy saving services: tunneling for fast rerouting, smooth node disabling and detection of both traffic spikes and link failures. We validate our solution by extensive simulations and by experimentation. We show that SENAtoR can be progressively deployed in a network using the SDN paradigm. It allows to reduce the energy consumption of ISP networks by 5 to 35% depending on the penetration of SDN hardware, while diminishing the packet loss rate compared to legacy protocols.

Software Defined Networking (SDN) is gaining momentum with the support of major manufacturers. While it brings flexibility in the management of flows within the data center fabric, this flexibility comes at the cost of smaller routing table capacities. In this paper, we investigate compression techniques to reduce the forwarding information base (FIB) of SDN switches. We validate our algorithm, called MINNIE, on a real testbed able to emulate a 20 switches fat tree architecture. We demonstrate that even with a small number of clients, the limit in terms of number of rules is reached if no compression is performed, increasing the delay of all new incoming flows. MINNIE, on the other hand, reduces drastically the number of rules that need to be stored with a limited impact on the packet loss rate. We also evaluate the actual switching and reconfiguration times and the delay introduced by the communications with the controller.

Software-Defined Networking (SDN) enables consolidation of the control plane of a set of network equipments with a fine-grained control of traffic flows inside the network. In this work, we demonstrate that some coarse-grained scheduling mechanisms can be easily offered by SDN switches without requiring any unsupported operation in OpenFlow. We leverage the feedback loop - flow statistics - exposed by SDN switches to the controller, combined with priority queuing mechanisms, usually available in typical switches on their output ports. We illustrate our approach through experimentations with an OpenvSwitch SDN switch controlled by a Beacon controller.