Network operators are facing significant challenges meeting the demand for more bandwidth, agile infrastructures, innovative services, while keeping costs low. Network Functions Virtualization (NFV) and Cloud Computing are emerging as key trends of 5G network architectures, providing flexibility, fast instantiation times, support of Commercial Off The Shelf hardware and significant cost savings. NFV leverages Cloud Computing principles to move the data-plane network functions from expensive, closed and proprietary hardware to the so-called Virtual Network Functions (VNFs). In this paper we deal with the management of virtual computing resources (Unikernels) for the execution of VNFs. This functionality is performed by the Virtual Infrastructure Manager (VIM) in the NFV MANagement and Orchestration (MANO) reference architecture. We discuss the instantiation process of virtual resources and propose a generic reference model, starting from the analysis of three open source VIMs, namely OpenStack, Nomad and OpenVIM. We improve the aforementioned VIMs introducing the support for special-purpose Unikernels and aiming at reducing the duration of the instantiation process. We evaluate some performance aspects of the VIMs, considering both stock and tuned versions. The VIM extensions and performance evaluation tools are available under a liberal open source licence.

Containers are in great demand because they are lightweight when compared to virtual machines. On the downside, containers offer weaker isolation than VMs, to the point where people run containers in virtual machines to achieve proper isolation. In this paper, we examine whether there is indeed a strict tradeoff between isolation (VMs) and efficiency (containers). We find that VMs can be as nimble as containers, as long as they are small and the toolstack is fast enough.

Video streaming dominates the Internet's overall traffic mix, with reports stating that it will constitute 90% of all consumer traffic by 2019. Most of this video is delivered by Content Delivery Networks (CDNs), and, while they optimize QoE metrics such as buffering ratio and start-up time, no single CDN provides optimal performance. In this paper we make the case for elastic CDNs, the ability to build virtual CDNs on-the-fly on top of shared, third-party infrastructure at a scale. To bring this idea closer to reality we begin by large-scale simulations to quantify the effects that elastic CDNs would have if deployed, and build and evaluate MiniCache, a specialized, minimalistic virtualized content cache that runs on the Xen hypervisor. MiniCache is able to serve content at rates of up to 32 Gb/s and handle up to 600K reqs/sec on a single CPU core, as well as boot in about 90 milliseconds on x86 and around 370 milliseconds on ARM32.

Increasing network speeds challenge the packet processing performance of networked systems. This can mainly be attributed to processing overhead caused by the split between the kernel-space network stack and user-space applications. To mitigate this overhead, we propose Santa, an application agnostic kernel-level cache of frequent requests. By allowing user-space applications to offload frequent requests to the kernel-space, Santa offers drastic performance improvements and unlocks the speed of kernel-space networking for legacy server software without requiring extensive changes.

Bit errors regularly occur in wireless communications. While many media streaming codecs in principle provide bit error tolerance and resilience, packet-based communication typically drops packets that are not transmitted perfectly. We present PICCETT, a method to heuristically identify which connections corrupted packets belong to, and to assign them to the correct applications instead of dropping them. PICCETT is a receiver-side classifier that requires no support from the sender or network, and no information which communication protocols are used. We show that PICCETT can assign virtually all packets to the correct connections at bit error rates up to 7-10%, and prevents misassignments even during error bursts. PICCETT's classification algorithm needs no prior offline training and both trains and classifies fast enough to easily keep up with IEEE 802.11 communication speeds.

Classical 802.11 rate adaptation algorithms rely on feedback from the receiver to correctly choose a sending rate, typically in the form of acknowledgments (ACKs). In the absence of such frames, novel techniques are required for rate selection. We present a novel On-Demand Feedback Rate Adaptation algorithm (OFRA) that works with ACK-less traffic. Feedback information is sent on-demand using a control frame to explicitly inform the transmitter about which bit rate to use on subsequent data frames. This approach guarantees standard conformity and exhibits fast and accurate bit rate adaptation at the cost of a modest overhead increase. We evaluate the performance of OFRA against various state-of-the-art rate adaptation schemes by means of simulations. If ACK frames are to be transmitted, OFRA performs better than related work in most considered scenarios, and on par in the others. In the absence of ACKs, OFRA provides large goodput gains under good channel conditions and comparable goodput in other situations.

Applications and protocols for wireless and mobile systems have to deal with volatile environmental conditions such as interference, packet loss, and mobility. Utilizing cross-layer information from other protocols and system components such as sensors can improve their performance and responsiveness. However, application and protocol developers lack a convenient way of monitoring, experimenting and specifying optimizations to evaluate their cross-layer ideas. We present CRAWLER, a novel experimentation architecture for system monitoring and cross-layer-coordination that facilitates evaluation of applications and wireless protocols. It alleviates the problem of complicated access to relevant system information by providing a unified interface for accessing application, protocol and system information. The generic design of this interface further enables a convenient and declarative way to specify and experiment how a set of cross-layer optimizations should be composed and adapted at runtime. Our evaluation demonstrates the usability of CRAWLER by experimenting, monitoring and improving TCP's congestion control algorithm.

High bit error rates reduce the performance of wireless networks. This is exacerbated by the enforcement of bit-by-bit correct transmissions and the resulting retransmission overhead. Recently, research has focused on more efficient link layer mechanisms and on tolerating payload errors. Header errors, however, still cause today's network and transport protocols to drop the erroneous packets.