Saturday, June 13, 2009
Topic/Presenter |
---|
Full AbstractA completely practical step by step guide to configuring IPv6 in a production network without breaking anything, for people who hate IPv6 and don't have it deployed currently. The goal would be that everyone who leaves the room should be able to successfully deploy an IPv6 network on top of an existing v4 network on any common Juniper/Cisco hardware, without needing to be IPv6 experts or lovers of the protocol. I also want to talk about practical techniques for addressing management so you can successfully deploy IPv6 without having successfully rewritten all of your internal management tools to fully support it, and other tricks to minimize the pain. Speakers |
Full AbstractThe IETF OPS & MGMT Area documents management technologies and operational best common practices. The purpose of this BoF is to review activities in that area and solicit feedback to determine the usefulness of those activities to the operator community. We will also solicit proposals for new work that is of interest to users. Speakers |
Sunday, June 14, 2009
Topic/Presenter |
---|
Full AbstractThe network capacity purchasing RFP can be a powerful tool that insures all options are on the table during pricing negotiations. It also allows the purchaser to remain firmly in control of the process. We aren't buying used cars, we're subscribing to radically changing costs that need to be managed closely. In this talk we will take a mid level walk through a tested RFP and discuss different aspects of an effective request for pricing or proposal "RFP". We will discuss how defining needs over multiple years can be strategically related to discounts and pricing, how to ask providers to define their own service delivery needs so you can recoup costs and standard terms and conditions to minimize lawyer time and expense. We'll conclude with a walk through of an actual RFP that will be provided as a template for later use. Speakers |
Full AbstractPresenting real-world experiences of effective outbound load-balancing across multiple ISPs using BGP traffic engineering and what we call "the metric system." Primary audience is content or data hosting networks that connect to multiple ISPs. Focus is on actual techniques that have been used successfully in numerous installations for simple, effective, and reliable load balancing. This tutorial will show in-depth and specific configurations to achieve desired traffic engineering and will share real-world results Speakers |
RecordingsFull AbstractAll major Cable Service Providers in United States have started the deployment of DOCSIS 3.0 at various scales. The objective of this tutorial is to provide a high level overview of the key components of DOCSIS3.0 technology. Speakers |
RecordingsFull AbstractSpeakers |
|
RecordingsFull AbstractThis presentation covers the topic of rapid convergence within IP networks. IP networks are frequently stated to not be capable of providing rapid convergence similar to that of SONET transport networks. However, with the right knob turns and a properly designed network, you can reach that 50ms or at least sub-second convergence target. The presentation will outline what applications require such performance, what protocols can deliver the results and how well it actually works in production. Additional lessons learned will be outlined showing operators that its not as simple as flipping on a switch. A list of commands and knobs are given at the end of the presentation to get operators a starting point to improve performance within their own networks. Speakers |
|
Full AbstractSpeakers |
RecordingsFull AbstractSpeakers |
Full AbstractMuch has been written about prefix hijacking but almost nothing has been presented and documented about best practices for preparing for a hijacking, detecting one, and responding when one happens. Speakers Tom Daly, Dynamic Network Services Anton Kapela, Voxel Todd Underwood, Google |
Monday, June 15, 2009
Topic/Presenter |
---|
Full AbstractDHS's National Infrastructure Protection Plan (NIPP) provides the coordinated approach that is used to establish national priorities, goals, and requirements for critical infrastructure and key resources (CI/KR) protection so that Federal funding and resources are applied in the most effective manner to reduce vulnerability, deter threats, and minimize the consequences of attacks and other incidents. It establishes the overarching concepts relevant to all CI/KR sectors identified in Homeland Security Presidential Directive-7 (HSPD-7), and addresses the physical, cyber, and human considerations required for effective implementation of protective programs. The NIPP calls for each sector (Information Technology and Communications are two of the eighteen sectors) to work with DHS through Sector Coordinating Councils (SCCs). This talk will provide a short overview of the NIPP, what the IT-SCC and Comm-SCC do, and how you can participate. Speakers |
Full AbstractThe ISP Security BOF aims to bring together operational security practitioners, vendors, and security researchers, to discuss BCPs, seasonable security topics, and other security, network engineering and operations related issues - with due focus on security. It also provides an opportunity for folks involved with operational security response to engage with others in the community, ask questions of the aggregate, introduce themselves and network with peers. Speakers Speaker - Randy Bush, IIJ Speaker - Warren Kumari, Google Speaker - Chris Morrow, Google. |
RecordingsFull AbstractThis track will include: Speakers Speaker - John Jason Brzozowski, Comcast Speaker - Randy Bush, IIJ Speaker - Tom Coffeen, Limelight Networks Speaker - Andy Davidson, LONAP/NetSumo Speaker - Dave Temkin, Netflix |
Full AbstractWhile various forms of application-layer peering overlays have long existed on the Internet to support applications like email and newsgroups, Voice over IP (VoIP) peering brings with it a new set of elements, policies, and challenges. This tutorial begins with the fundamentals, exploring why peering is even necessary for a largely client-driven application like VoIP. It then surveys the functional elements necessary to peer between telephony-replacing deployments, using the ongoing IETF work on VoIP peering as a touchstone. All of this is considered with a mind to the ongoing evolution of traditional voice services, including mobile telephony, and the rise of other client-driven Internet applications including instant messaging and peer-to-peer file sharing. Speakers |
Full AbstractAlcatel-Lucent |
Full AbstractNewcomer's Continental Breakfast Speakers |
RecordingsFull AbstractSpeakers |
RecordingsFull AbstractThe internet has made many things easier in human society, especially including crime. For the last decade the good guys keep losing and the bad guys keep getting rich -- why? Drawing from contemporary examples including the 2008/2009 "conficker worm", from the rise of drug-resistant microbes, and from Sun Tzu's 2,500-year-old "The Art of War", I will try to explain how our approach and our philosophy defeats us even though we remain the superior force. Speakers |
Full AbstractComcast |
RecordingsFull AbstractThis talk will address the issue with an increased number of unique routes in the routing system, and discuss how the number of routes is increasing more steeply than the number of prefixes. It will discuss how external and internal network architecture has effects on the number of routes, and how a number of other components are effected as a result. It will specifically discuss internal BGP scaling issues, e.g., that of route reflection, and some improvements that can be found by modifying network routing design, implementations, and the BGP protocol itself. Speakers |
Full AbstractThe presentation is Part II in the series (1st reference, NANOG43) where we focus on: current business case of why the WISP business is still relevant, wireless technology, the new challenges/benefits associated with WiMax Speakers |
RecordingsFull AbstractSpeakers Panelist - Richard Donaldson, 6Connect Panelist - Tesh Durvasala, Telx Panelist - Lane Patterson, Equinix Panelist - Dave Pickut, Equinix Josh Snowhorn, Terremark Ben Stewart, Terremark Worldwide |
|
Full AbstractSpeakers |
Full AbstractSpeakers |
Full AbstractSpeakers |
Full AbstractSoftware bugs in routers lead to network outages, security vulnerabilities, and other unexpected behavior. Rather than simply crashing the router, bugs can violate protocol semantics, rendering traditional failure detection and recovery techniques ineffective. Handling router bugs is an increasingly important problem as new applications demand higher availability, and networks become better at dealing with traditional failures. Further demonstrating the importance is a string of recent high profile outages, including a very recent incident where a single prefix announcement to a single provider caused a huge increase in the global update rate and instability due to two bugs in routers from two different vendors. In this paper, we tailor software and data diversity (SDD) to the unique properties of routing protocols, to avoid buggy behavior at run time. Our bug-tolerant router executes multiple diverse instances of routing software, and uses voting to determine the output to publish to the forwarding table, or to advertise to neighbors. We design and implement a router hypervisor that makes this parallelism transparent to other routers, handles fault detection and booting of new router instances, and performs voting in the presence of routing-protocol dynamics, without need to modify software of the diverse instances. Experiments with BGP message traces and the XORP and Quagga open-source software running on our Linux-based router hypervisor demonstrate that our solution scales to large networks and efficiently masks buggy behavior. Speakers Jennifer Rexford, Princeton University Minlan Yu, Princeton University. |
Full AbstractSpeakers |
Tuesday, June 16, 2009
Topic/Presenter |
---|
Full AbstractSpeakers |
Full AbstractSpeakers |
Full AbstractIntroduction - Ryan Donnelly, VeriSign Speakers |
Full AbstractThis session is intended to provide an introduction to DHCPv6. DHCPv4 and DHCPv6 are separate and distinct protocols. DHCPv6 will be compared and contrast to DHCPv4, specific topics include techniques that can be used to offer redundant DHCPv6 services in the absence of a standards based protocol. Additional topics related to the evolution and future use of DHCPv6 will also be discussed. Speakers |
Full AbstractWhat feature mix can we expect from .1 to ~.5Tb/s top of the rack sized l3 switches. Speakers |
Full AbstractSpeakers |
|
Full AbstractSpeakers |
Full AbstractEquinix |
Full AbstractPseudowires where invented during the latest Tech boom in late 1999 out of necessity of replacing ATM networks with a more efficient technology. The technology quickly became very popular, and Speakers |
Full AbstractAfter a dozen or so years of protocol development, operational whiteboarding, and implementation effort, the Security Extensions to Speakers |
RecordingsFull AbstractSpeakers |
Full AbstractThere exist many top-of-rack switch redundancy models allowing for resilient access to servers in the datacenter. In this presentation various configurations will be demonstrated along with benefits, drawbacks, and real-world results. Information provided should assist datacenter network operators with selecting the right redundancy model for their application and environment. Speakers |
Wednesday, June 17, 2009
Topic/Presenter |
---|
Full AbstractAlcatel-Lucent |
RecordingsFull AbstractSpeakers |
RecordingsFull AbstractSpeakers |
Full AbstractSpeakers |
RecordingsFull AbstractSpeakers |
Full AbstractIn our presentation, we would like to share the experience gained by carrying out the early phases of IPv6 testing and limited deployment in the Greek Student Network (Diodos). Diodos is a state sponsored ADSL access network that is operated by the Greek Research Network and provides network access to academic users across the country. With 50 institutions involved and more than 20000 academic users connected, assigning IPv6 prefixes to each user according to the respective institution presents some interesting addressing and routing problems, which we describe in detail and propose our solution. Furthermore, we describe the use of PPPv6, DHCPv6 prefix delegation or other possible options to enumerate the LAN behind each user CPE and we elaborate on RADIUS AV pairs, accounting problems, DNS options, ACLs, implementation availability, critical bugs and various miscellaneous details for Cisco based equipment. Lastly, we describe the state of our testing deployment and lay out the problems that are critical and must be addressed in the future. Speakers |
|
|
|
RecordingsFull AbstractAddress family translation is being discussed as a viable strategy to deal with IPv4 address exhaustion and IPv6 transition. In the former sense, it can be used to allocate private addresses to customers of a service provider as the NAT is under the provider's control. In the V6 transition case it is seen as a method for either allocating V6 addresses to subscribers when the provider network isn't fully V6 enabled or when a provider's network is V6 enabled and a subscriber is not V6 ready. No matter how one view's NATs there is renewed interest in deploying much larger NATs that perform multiple functions. There are a couple of critical items in this round of deployment: they may be deployed under the provider's control and there may be conjoined encapsulation technology. Speakers Panelist - Chris Chase, AT&T Labs Panelist - Alain Durand, Comcast Panelist - Suzanne Woolf, Internet Systems Consortium. |
RecordingsFull AbstractSpeakers |