Adam Greenhalgh

Improving datacenter performance and robustness with multipath TCP

The latest large-scale data centers offer higher aggregate bandwidth and robustness by creating multiple paths in the core of the net- work. To utilize this bandwidth requires different flows take different paths, which poses a challenge. In short, a single-path transport seems ill-suited to such networks. We propose using Multipath TCP as a replacement for TCP in such data centers, as it can effectively and seamlessly use available bandwidth, giving improved throughput and better fairness on many topologies. We investigate what causes these benefits, teasing apart the contribution of each of the mechanisms used by MPTCP. Using MPTCP lets us rethink data center networks, with a different mindset as to the relationship between transport protocols, rout- ing and topology. MPTCP enables topologies that single path TCP cannot utilize. As a proof-of-concept, we present a dual-homed variant of the FatTree topology. With MPTCP, this outperforms FatTree for a wide range of workloads, but costs the same. In existing data centers, MPTCP is readily deployable leveraging widely deployed technologies such as ECMP. We have run MPTCP on Amazon EC2 and found that it outperforms TCP by a factor of three when there is path diversity. But the biggest benefits will come when data centers are designed for multipath transports.

Network virtualization architecture: proposal and initial prototype

The tussle between reliability and functionality of the Internet is firmly biased on the side of reliability. New enabling technologies fail to achieve traction across the majority of ISPs. We believe that the greatest challenge is not in finding solutions and improvements to the Internets many problems, but in how to actually deploy those solutions and re-balance the tussle between reliability and functionality. Network virtualization provides a promising approach to enable the coexistence of innovation and reliability. We describe a network virtualization architecture as a technology for enabling Internet innovation. This architecture is motivated from both business and technical perspectives and comprises four main players. In order to gain insight about its viability, we also evaluate some of its components based on experimental results from a prototype implementation.

Towards high performance virtual routers on commodity hardware

Modern commodity hardware architectures, with their multiple multi-core CPUs and high-speed system interconnects, exhibit tremendous power. In this paper, we study performance limitations when building both software routers and software virtual routers on such systems. We show that the fundamental performance bottleneck is currently the memory system, and that through careful mapping of tasks to CPU cores, we can achieve forwarding rates of 7 million minimum-sized packets per second on mid-range server-class systems, thus demonstrating the viability of software routers. We also find that current virtualisation systems, when used to provide forwarding engine virtualisation, yield aggregate performance equivalent to that of a single software router, a tenfold improvement on current virtual router platform performance. Finally, we identify principles for the construction of high-performance software router systems on commodity hardware, including full router virtualisation support.

Steps towards a DoS-resistant internet architecture

Defending against DoS attacks is extremely difficult; effective solutions probably require significant changes to the Internet architecture. We present a series of architectural changes aimed at preventing most flooding DoS attacks, and making the remaining attacks easier to defend against. The goal is to stimulate a debate on trade-offs between the flexibility needed for future Internet evolution and the need to be robust to attack.

Evaluating Xen for Router Virtualization

In this paper, we evaluate the performance of a software IP router forwarding plane inside the Xen virtual machine monitor environment with a view to identifying (some) design issues in Virtual Routers. To this end, we evaluate and compare the forwarding performance of two identical Linux software router configurations, run either above the Xen hypervisor or within vanilla Linux. Even with minimal sized packets, we show that the Xen DomO privileged domain offers near native forwarding performance at the condition that the sollicitation to unpriviledged domains stay minimal, whereas Xen unprivileged domains offer very poor performance in every cases. This shows that an important design principle for virtual router platforms must be to handle all forwarding, for all virtual routers, onto the same forwarding engine, in order to avoid much detrimental per-packet context switching.

Data center networking with multipath TCP

Recently new data center topologies have been proposed that offer higher aggregate bandwidth and location independence by creating multiple paths in the core of the network. To effectively use this bandwidth requires ensuring different flows take different paths, which poses a challenge. Plainly put, there is a mismatch between single-path transport and the multitude of available network paths. We propose a natural evolution of data center transport from TCP to multipath TCP. We show that multipath TCP can effectively and seamlessly use available bandwidth, providing improved throughput and better fairness in these new topologies when compared to single path TCP and randomized flow-level load balancing. We also show that multipath TCP outperforms laggy centralized flow scheduling without needing centralized control or additional infrastructure.

Improved Forwarding Architecture and Resource Management for Multi-Core Software Routers

Recent technological advances in commodity server architectures, with multiple multi-core CPUs, integrated memory controllers, high-speed interconnects and enhanced network interface cards, provide substantial computational capacity and thus an attractive platform for packet forwarding. However, to exploit this available capacity, we need a suitable software platform that allows effective parallel packet processing and resource management. In this paper, we at first introduce an improved forwarding architecture for software routers that enhances parallelism by exploiting hardware classification and multi-queue support, already available in recent commodity network interface cards. After evaluating the original scheduling algorithm of the widely-used Click modular router, we propose solutions for extending this scheduler for improved fairness, throughput and more precise resource management. To illustrate the potential benefits of our proposal, we implement and evaluate a few key elements of our overall design.

A platform for high performance and flexible virtual routers on commodity hardware

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.

Forwarding path architectures for multicore software routers

Multi-core CPUs, along with recent advances in memory and buses, render commodity hardware a strong candidate for building fexible and high-performance software routers. With a forwarding plane physically composed of many packet processing components and operations, resource allocation in multi-core systems is not trivial. Indeed, packets crossing cache hierarchies degrade forwarding performance, since the bottleneck is main memory access. Therefore, forwarding path allocation and input/output processing become challenging, especially when states and data structures have to be shared among multiple cores. In this context, we investigate a set of input/output processing architectures, as well as resource allocation strategies for forwarding paths. For each packet processing operation, we uncover the gains and possible implications by either running different components concurrently or replicating the same components across different cores.

Designing a Platform for Flexible and Performant Virtual Routers on Commodity Hardware

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 x86 hardware. We also elaborate on our design choices in order to achieve both high performance and flexibility for packet processing.