Search

This article summarises the presentations and discussions during a workshop on end-to-end protocols for the future Internet in June 2008. The aim of the workshop was to establish a dialogue at the interface between two otherwise fairly distinct communities working on future Internet protocols: those developing internetworking functions and those developing end-to-end transport protocols. The discussion established near-consensus on some of the open issues, such as the preferred placement of traffic engineering functionality, whereas other questions remained controversial. New research agenda items were also identified.

The Internet has seen a proliferation of specialized middlebox devices that carry out crucial network functionality such as load balancing, packet inspection and intrusion detection. Recent advances in CPU power, memory, buses and network connectivity have turned commodity PC hardware into a powerful network platform. Furthermore, commodity switch technologies have recently emerged offering the possibility to control the switching of flows in a fine-grained manner. Exploiting these new technologies, we present a new class of network architectures which enables flow processing and forwarding at unprecedented flexibility and low cost.

Defending against large, distributed Denial-of-Service attacks is challenging, with large changes to the network core or to end-hosts often suggested. To make matters worse, spoofing adds to the difficulty, since defenses must resist attempts to trigger filtering of other people's traffic. Further, any solution has to provide incentives for deployment, or it will never see the light of day. We present a simple and effective architectural defense against distributed DoS attacks that requires no changes to the end-hosts, minimal changes to the network core, is robust to spoofing, provides incentives for initial deployment, and can be built with off-the-shelf hardware

The core Internet protocols have not changed significantly in more than a decade, in spite of exponential growth in the number of Internet users and the speed of the fastest links. The requirements placed on the net are also changing, as digital convergence finally occurs. Will the Internet cope gracefully with all this change, or are the cracks already beginning to show? In this paper I examine how the Internet has coped with past challenges resulting in attempts to change the architecture and core protocols of the Internet. Unfortunately, the recent history of failed architectural changes does not bode well. With this history in mind, I explore some of the challenges currently facing the Internet.

Questioning whether layering is still an adequate foundation for networking architectures, this paper investigates non-layered approaches to the design and implementation of network protocols. The goals are greater flexibility and control with fewer feature interaction problems. The paper further proposes a specific non-layered paradigm called role-based architecture .

Network researchers face a significant problem when deploying software in routers, either for experimentation or for pilot deployment. Router platforms are generally not open systems, in either the open-source or the open-API sense. In this paper we discuss the problems this poses, and present an eXtensible Open Router Platform (XORP) that we are developing to address these issues. Key goals are extensibility, performance and robustness. We show that different parts of a router need to prioritize these differently, and examine techniques by which we can satisfy these often conflicting goals. We aim for XORP to be both a research tool and a stable deployment platform, thus easing the transition of new ideas from the lab to the real world.

Contemporary network stacks are masterpieces of generality, supporting many edge-node and middle-node functions. Generality comes at a high performance cost: current APIs, memory models, and implementations drastically limit the effectiveness of increasingly powerful hardware. Generality has historically been required so that individual systems could perform many functions. However, as providers have scaled services to support millions of users, they have transitioned toward thousands (or millions) of dedicated servers, each performing a few functions. We argue that the overhead of generality is now a key obstacle to effective scaling, making specialization not only viable, but necessary. We present Sandstorm and Namestorm, web and DNS servers that utilize a clean-slate userspace network stack that exploits knowledge of application-specific workloads. Based on the netmap framework, our novel approach merges application and network-stack memory models, aggressively amortizes protocol-layer costs based on application-layer knowledge, couples tightly with the NIC event model, and exploits microarchitectural features. Simultaneously, the servers retain use of conventional programming frameworks. We compare our approach with the FreeBSD and Linux stacks using the nginx web server and NSD name server, demonstrating 2--10x and 9x improvements in web-server and DNS throughput, lower CPU usage, linear multicore scaling, and saturated NIC hardware.

Streaming audio and video applications are becoming increasingly popular on the Internet, and the lack of effective congestion control in such applications is now a cause for significant concern. The problem is one of adapting the compression without requiring video servers to reencode the data, and fitting the resulting stream into the rapidly varying available bandwidth. At the same time, rapid fluctuations in quality will be disturbing to the users and should be avoided. We present a mechanism for using layered video in the context of unicast congestion control. This quality adaptation mechanism adds and drops layers of the video stream to perform long-term coarse-grain adaptation, while using a TCP-friendly congestion control mechanism to react to congestion on very short timescales. The mismatches between the two timescales are absorbed using buffering at the receiver. We present an efficient scheme for the distribution of available bandwidth among the active layers. Our scheme allows the server to trade short-term improvement for long-term smoothing of quality. We discuss the issues involved in implementing and tuning such a mechanism, and present our simulation results.

To search the web quickly, search engines partition the web index over many machines, and consult every partition when answering a query. To increase throughput, replicas are added for each of these machines. The key parameter of these algorithms is the trade-off between replication and partitioning: increasing the partitioning level improves query completion time since more servers handle the query, but may incur non-negligible startup costs for each sub-query. Finding the right operating point and adapting to it can significantly improve performance and reduce costs. We introduce Rendezvous On a Ring (ROAR), a novel distributed algorithm that enables on-the-fly re-configuration of the partitioning level. ROAR can add and remove servers without stopping the system, cope with server failures, and provide good load-balancing even with a heterogeneous server pool. We demonstrate these claims using a privacy-preserving search application built upon ROAR.

Since the ARPAnet, network designers have built localized mechanisms for statistical multiplexing, load balancing, and failure resilience, often without understanding the broader implications. These mechanisms are all types of resource pooling, whichmeans making a collection of resources behave like a single pooled resource. We believe that the natural evolution of the Internet is that it should achieve resource pooling by harnessing the responsiveness of multipath-capable end systems. We argue that this approach will solve the problems and limitations of the current piecemeal approaches. This research was supported by Trilogy (http://www.trilogy-project.org), a research project (ICT-216372) partially funded by the European Community under its Seventh Framework Programme. European Community's Seventh Framework Program Publicado

Fast-growing Internet applications like streaming media and telephony prefer timeliness to reliability, making TCP a poor fit. Unfortunately, UDP, the natural alternative, lacks congestion control. High-bandwidth UDP applications must implement congestion control themselves-a difficult task-or risk rendering congested networks unusable. We set out to ease the safe deployment of these applications by designing a congestion-controlled unreliable transport protocol . The outcome, the Datagram Congestion Control Protocol or DCCP, adds to a UDP-like foundation the minimum mechanisms necessary to support congestion control. We thought those mechanisms would resemble TCP's, but without reliability and, especially, cumulative acknowledgements, we had to reconsider almost every aspect of TCP's design. The resulting protocol sheds light on how congestion control interacts with unreliable transport, how modern network constraints impact protocol design, and how TCP's reliable bytestream semantics intertwine with its other mechanisms, including congestion control.

Multi-core CPUs, along with recent advances in memory and buses, render commodity hardware a strong candidate for software router virtualization. In this context, we present the design of a new platform for virtual routers on modern PC hardware. We further discuss our design choices in order to achieve both high performance and flexibility for packet processing.

This paper proposes a mechanism for equation-based congestion control for unicast traffic. Most best-effort traffic in the current Internet is well-served by the dominant transport protocol, TCP. However, traffic such as best-effort unicast streaming multimedia could find use for a TCP-friendly congestion control mechanism that refrains from reducing the sending rate in half in response to a single packet drop. With our mechanism, the sender explicitly adjusts its sending rate as a function of the measured rate of loss events, where a loss event consists of one or more packets dropped within a single round-trip time. We use both simulations and experiments over the Internet to explore performance.

In this paper we provide an overview of multimedia conferencing on the Internet. The protocols mentioned are all specified elsewhere as Internet-Drafts or RFCs. Each RFC gives details of the protocol itself, how it works and what it does. This document attempts to provide the reader with an overview of how the components fit together and of some of the assumptions made, as well as some statement of direction for those components still in a nascent stage.