In this paper, we quantify reordering and connection interruption while downloading data to mobile phones in modern HSPA/HSPA+ mobile networks based on end-to-end measurements conducted in commercial operator networks. We show how these events are affected by phone model, operator, time of the day, user movement and signal strength. We find that phone model, time of day and user movement affect the number of connection interruptions, but not their length. Data loss during interruptions is influenced by operator and time of day. The number of reordering events is impacted by the operator and the time of day, the operator additionally affects the reordering extent. This improves the understanding of HSPA/HSPA+ networks and enables further research to improve the data transfer and methods for delay spike prevention.

Bloom-filter-based forwarding has been suggested to solve several fundamental problems in the current Internet, such as routing-table growth, multicast scalability issues, and denial-of-service (DoS) attacks by botnets. The proposed protocols are source-routed and include the delivery tree encoded as a Bloom filter in each packet. The network nodes forward packets based on this in-packet information without consulting routing tables and without storing per-flow state. We show that these protocols have critical vulnerabilities and make several false security assumptions. In particular, we present DoS attacks against broad classes of Bloom-filter-based protocols and conclude that the protocols are not ready for deployment on open networks. The results also help us understand the limitations and design options for Bloom-filter forwarding.

Ethernet has become the de facto layer 2 transmission technology, partly due to ease of use and cost efficiency. The cores of most data centers around the world are based on Ethernet, and large access and core networks are built without IP routing. The inherent simplicity of Ethernet has several drawbacks, including the overhead of network-wide broadcasting and the use of the spanning tree protocol. The academia and the industry have presented numerous solutions to overcome the limitations of Ethernet. Many of these solutions have deployment constraints and often force a change of the whole network. In this paper, we present DBridges, an efficient solution for building large-scale Ethernets based on the IETF-driven RBridges and SEATTLE. DBridges provide loop-safety, dramatic decrease in broadcast traffic, and resiliency and graceful failover without affecting the plug-and-play vision of Ethernet. The key contributions of our solution are that we are backward compatible with RBridges and can support incremental deployment. We analyze and show these benefits in detail, and present various performance measurements using our proof-of-concept implementation. We estimate that with DBridges it is possible to increase the size of Ethernet broadcast domains at least tenfold.

Several recently proposed multicast protocols use in-packet Bloom filters to encode multicast trees. These mechanisms are in principle highly scalable because no per-flow state is required in the routers and because routing decisions can be made efficiently by simply checking for the presence of outbound links in the filter. Yet, the viability of previous approaches is limited by the possibility of forwarding anomalies caused by false positives inherent in Bloom filters. This paper explores such anomalies, namely (1) packets storms, (2) forwarding loops and (3) flow duplication. We propose stateless solutions that increase the robustness and the scalability of Bloom filter-based multicast protocols. In particular, we show that the parameters of the filter need to be varied to guarantee the stability of the packet forwarding, and we present a bit permutation technique that effectively prevents both accidental and maliciously created anomalies. We evaluate our solutions in the context of BloomCast, a source-specific inter-domain multicast protocol, using analytical methods and simulations.

Several new, data-oriented internetworking architectures have been proposed recently. However, the practical deployability of such designs is an open question. In this paper, we consider data-oriented network designs in the light of the policy and incentive structures of the present internetworking economy. A main observation of our work is that none of the present proposals is both policy-compliant and incentive-compatible with the current internetworking market, which makes their deployment very challenging if not impossible. This difficulty stems from the unfounded implicit assumption that data-oriented routing policies directly reflect the underlying packet-level inter-domain policies. We find that to enable the more effective network utilization promised by data-oriented networking, essential caching incentives need to exist, and that data-oriented peering needs be considered separately from peering for packet forwarding.