Showing papers by "Anja Feldmann published in 2007"

Live wide-area migration of virtual machines including local persistent state

Abstract: So far virtual machine (VM) migration has focused on transferring the run-time memory state of the VMs in local area networks (LAN). However, for wide-area network (WAN) migration it is crucial to not just transfer the VMs image but also transfer its local persistent state (its file system) and its on-going network connections. In this paper we address both: by combining a block-level solution with pre-copying and write throttling we show that we can transfer an entire running web server, including its local persistent state, with minimal disruption --- three seconds in the LAN and 68 seconds in the WAN); by combining dynDNS with tunneling, existing connections can continue transparently while new ones are redirected to the new network location. Thus we show experimentally that by combining well-known techniques in a novel manner we can provide system support for migrating virtual execution environments in the wide area.

Can ISPS and P2P users cooperate for improved performance

Abstract: Peer-to-peer (P2P) systems, which are realized as overlays on top of the underlying Internet routing architecture, contribute a significant portion of today's Internet traffic. While the P2P users are a good source of revenue for the Internet Service Providers (ISPs), the immense P2P traffic also poses a significant traffic engineering challenge to the ISPs. This is because P2P systems either implement their own routing in the overlay topology or may use a P2P routing underlay [1], both of which are largely independent of the Internet routing, and thus impedes the ISP's traffic engineering capabilities. On the other hand, P2P users are primarily interested in finding their desired content quickly, with good performance. But as the P2P system has no access to the underlying network, it either has to measure the path performance itself or build its overlay topology agnostic of the underlay. This situation is disadvantageous for both the ISPs and the P2P users.To overcome this, we propose and evaluate the feasibility of a solution where the ISP offers an "oracle" to the P2P users. When the P2P user supplies the oracle with a list of possible P2P neighbors, the oracle ranks them according to certain criteria, like their proximity to the user or higher bandwidth links. This can be used by the P2P user to choose appropriate neighbors, and therefore improve its performance. The ISP can use this mechanism to better manage the immense P2P traffic, e.g., to keep it inside its network, or to direct it along a desired path. The improved network utilization will also enable the ISP to provide better service to its customers.

Internet clean-slate design: what and why?

Abstract: Many believe that it is impossible to resolve the challenges facing today's Internet without rethinking the fundamental assumptions and design decisions underlying its current architecture. Therefore, a major research effort has been initiated on the topic of Clean Slate Design of the Internet's architecture. In this paper we first give an overview of the challenges that a future Internet has to address and then discuss approaches for finding possible solutions, including Clean Slate Design. Next, we discuss how such solutions can be evaluated and how they can be retrofitted into the current Internet. Then, we briefly outline the upcoming research activities both in Europe and the U. S. Finally, we end with a perspective on how network and service operators may benefit from such an initiative.

Packet capture in 10-gigabit Ethernet environments using contemporary commodity hardware

Abstract: Tracing traffic using commodity hardware in contemporary high-speed access or aggregation networks such as 10-Gigabit Ethernet is an increasingly common yet challenging task. In this paper we investigate if today's commodity hardware and software is in principle able to capture traffic from a fully loaded Ethernet. We find that this is only possible for data rates up to 1 Gigabit/s without reverting to using special hardware due to, e. g., limitations with the current PC buses. Therefore, we propose a novel way for monitoring higher speed interfaces (e.g., 10-Gigabit) by distributing their traffic across a set of lower speed interfaces (e.g., 1-Gigabit). This opens the next question: which system configuration is capable of monitoring one such 1-Gigabit/s interface? To answer this question we present a methodology for evaluating the performance impact of different system components including different CPU architectures and different operating system. Our results indicate that the combination of AMD Opteron with FreeBSD outperforms all others, independently of running in single- or multi-processor mode. Moreover, the impact of packet filtering, running multiple capturing applications, adding per packet analysis load, saving the captured packets to disk, and using 64-bit OSes is investigated.

Enabling seamless internet mobility

Abstract: Mobility is a requirement not appropriately addressed by the original design of the Internet. A plethora of suggestions have been made to overcome this.We propose the Seamless Internet Mobility System (SIMS) for enabling seamless IP network layer mobility. SIMS is incrementally deployable in today's IPv4 based Internet. Contrary to other mobility solutions (e.g., MIP), it adds little overhead and can be used even without a permanent IP address or a home agent.

Automated Parallel Solution of Unstructured PDE Problems

Abstract: Xerox CorporationPalo Alto Research Center3333 Coyote Hill RoadPalo Alto, CA 94304CR Categories and Subject Descriptors: F.2.2 [Analysis of Algorithms and Problem Complex-ity]: Nonnumerical Algorithms and Problems; G.1.0 [Numerical Analysis]: General; G.2.2 [DiscreteMathematics]: Graph TheoryGeneral Terms: AlgorithmsAdditional Key Words and Phrases: Partial differential equations, finite element methods, graphpartitioning, parallel processing.