Saturday, October 2, 2010
Topic/Presenter
Full Abstract

How to get the best out of your attendance at NANOG. What on earth is a BoF or a Track? Famous faces, usual suspects. Go on, submit a lightning talk. If you like it tell your friends, if you don't, tell us.

Speakers
Mike Hughes, Program Committee
Mike Hughes was previously Chief Technical Officer for London Internet Exchange (LINX), where he was responsible for the organisation's overall technical strategy, core peering infrastructure, and operational performance, specialising in high-speed metro ethernet platforms. With over 10 years of industry experience, Mike has become involved in activities within the community as a regular participant at industry for a (such as NANOG and RIPE meetings), and is a co-chair of the RIPE European Internet Exchange Working Group, as well as being a member of the UKNOF programme committee. He also sits on the Customer Technical Advisory Council of Extreme Networks.

Full Abstract

The tutorial introduces service providers to some more advanced BGP features and techniques to aid with operating their networks within the Internet. After a recap of iBGP, eBGP and common attributes, the tutorial will look at the various scaling techniques available, when to use BGP instead of an IGP, and examine policy options available through the use of local preference, MED and communities. The second half of the tutorial looks at deployment techniques, including BGP network design, the announcing and receiving refixes, aggregation, routing table growth and stability, finishing off with some configuration advice.

Speakers
Philip Smith, Cisco Systems.

Full Abstract

IEEE 802.1aq Shortest Path Bridging (presented at Nanog 49) is a new layer 2 control protocol using IS-IS to replace and greatly enchance the capabilities of the layer 2 data paths Q-in-Q and Mac-in-Mac. The shortest and multi path virtualized routing have applications in the Data Center , Metro, and Enterprise we will give an overview of 802.1aq, then drop into a deeper dive into how it works, looking at some of the algorithms in some detail, and will finish with a look at some live CLI on a 30+ node network. An emulator will also be available and time permitting we can experiment with different networks to see the behaviors.

Speakers
Peter Ashwood-Smith, Huawei Technologies Canada
Peter Ashwood-Smith ([email protected]) is a Huawei Fellow with B.S and M.S. degrees in computer science from the Universities of Victoria and Toronto. He has worked on the design, standardization, implementation, deployment, and support of many modern routing protocols (e.g., label switched networks including PORS, ATM, MPLS, and GMPLS). Peter previously worked at Nortel where he held a number of product related positions and a fellowship. He currently has approximately 32 networking related patents.
His interests and research are now tending in the direction of computation-based/non-label-switched protocols and as a result Peter is now participating in the design, standardization, implementation and productization of IEEE 802.1aq/Shortest Path Bridging.

Full Abstract

With the mandated roll out of DNSSEC for US Government and the widespread adoption of DNSSEC throughout the world, many network operators are being asked to make changes to infrastructure that has been stable for years.

With the introduction of BIND 9.7 (featuring "DNSSEC for Humans"), ISC has attempted to remove much of the hassle from deploying DNSSEC.

While BIND 9.7 makes deployment easier, there are still hurdles to be overcome.

This presentation will provide attendees with information on changes to workflow and configurations that will allow a less painful deployment of DNSSEC using ISC's BIND.

Speakers
Alan Clegg, Internet Systems Consortium
Alan Clegg has over 20 years' experience providing support for and management of Internet-facing systems. As a Dale Carnegie–trained public speaker, Alan has provided tailored learning experiences to corporations and at conventions and meetings (BSDcon, InfraGard, HTCIA) around the globe. Since joining the Internet Systems Consortium staff in 2007, Alan has been creating and providing workshops and training for ISC customers and users. The training sessions include a 5-day DNS and BIND class, a 3-day DNSSEC workshop, and a 2-day ISC DHCP course.

Sunday, October 3, 2010
Topic/Presenter
Full Abstract

The second half of the tutorial looks at deployment techniques, including BGP network design, the announcing and receiving prefixes, aggregation, routing table growth and stability, finishing off with some configuration advice.

Speakers
Philip Smith, Cisco Systems.

Full Abstract

RBridges, based on the IETF TRILL protocol, provide safe optimal pair-wise forwarding without configuration and support for multipathing of both unicast and multicast traffic. They achieve this with IS-IS routing and encapsulation of traffic with a header that includes a hop count. RBridges are compatible with previous IEEE 802.1 customer bridges, as well as IPv4 and IPv6 routers and end nodes, and such bridges can be incrementally replaced by RBridges. Current work on minimizing broadcast traffic, including ARP/ND optimization, will also be covered.

Speakers
Donald Eastlake, Stellar Switches
Donald Eastlake is co-chair of the IETF TRILL Working Group and a member of the IEEE 802.1 Working Group. Donald is author of about 50 IETF RFCs, including the only RFC with "sex" in its title. He was also the editor of the TRILL base protocol specification, and an author of a number of other TRILL related Internet Drafts. Previous, he was a Distinguished Member of Technical Staff at Motorola.

Full Abstract

Speakers
Steve Feldman, Steering Committee Chair, CBS Interactive

Full Abstract

We demonstrate a low-cost MPLS LSR capable of forwarding 4x1GE in hardware. It utilizes an open-source implementation of LDP in Quagga, open-source modifications to the Linux kernel to support MPLS, an open-source implementation of an OpenFlow controller modified to support MPLS, and a NetFPGA card as the open platform to program the hardware for MPLS forwarding.

Speakers
Scott Whyte, Google

Full Abstract

We all see the size of the global routing table constantly growing, but what is really behind the number of prefixes being announced? A knee jerk answer is "more people on the internet", or "intentional deaggregation for policy reasons", but does this really explain it? How much of the routing table size is caused by unintentional configurations, or slow growth allocations? Can we do a better job of identifying these route sources by looking at common BGP attributes rather than simple route aggregation techniques (which often miss "holes" in otherwise aggregatable large blocks)? And is IPv6 really helping to improve this issue and reduce routing table size, or are people just making the same mistakes all over again.

Speakers
Richard Steenbergen, nLayer Communication

Monday, October 4, 2010
Topic/Presenter
Full Abstract

Two competing methods of solving the multipath/redundant L2 Ethernet challenge have been emerging over the years, and have now become viable. In this showdown of TRILL versus 802.1aq (SPB) we will debate issues including stability, convergence, vendor support, and standardization.

Speakers
Moderator - Dani Roisman, None
Dani Roisman is currently the Sr. Director of network engineering and Operations at The Planet Internet Services, where he oversees the teams that build and run the network that supports over 45,000 managed servers across 7 datacenters. Prior to The Planet, he was a Senior Network Architect and Network Engineering Team Lead at Peak Web Consulting, and Director of Network Engineering and Facilities Architecture at Sony Online Entertainment.

Dani has specialized in large content datacenters and networks, with a focus on peering and multi-homing to reduce costs, improve customer negotiating stance, as well as increased network capacity, performance, and fault-tolerance. His network design and implementation accomplishments include massively-multiplayer games (EverQuest, Star Wars Galaxies, and PlanetSide) and social networking deployments hosting over 5,000 servers across 9 datacenters with Internet bandwidth capacities reaching 250Gbps.

Panelist - Peter Ashwood-Smith, Huawei Fellow
Peter Ashwood-Smith ([email protected]) is a Huawei Fellow with B.S and M.S. degrees in computer science from the Universities of Victoria and Toronto. He has worked on the design, standardization, implementation, deployment, and support of many modern routing protocols (e.g., label switched networks including PORS, ATM, MPLS, and GMPLS). Peter previously worked at Nortel where he held a number of product related positions and a fellowship. He currently has approximately 32 networking related patents. His interests and research are now tending in the direction of computation-based/non-label-switched protocols and as a result Peter is now participating in the design, standardization, implementation and productization of IEEE 802.1aq/Shortest Path Bridging.

Panelist - Donald Eastlake, TRILL
Donald Eastlake is co-chair of the IETF TRILL Working Group and a member of the IEEE 802.1 Working Group. Donald is author of about 50 IETF RFCs, including the only RFC with "sex" in its title. He was also the editor of the TRILL base protocol specification, and an author of a number of other TRILL related Internet Drafts. Previous, he was a Distinguished Member of Technical Staff at Motorola.

Panelist - Srikanth Keesara, Avaya
Srikanth Keesara is an Architect for the Modular Ethernet product portfolio at Avaya. He has worked for the past few years on SPB based product development and has been actively engaged in the deployment and support of this technology in large customer networks. Srikanth worked at Nortel within the Metro Ethernet Networks division and participated in the development of products based on IEEE-802.1ah, IEEE-802.1ag and SPB. He is a graduate of Indian Institute of Technology, Mumbai and Indian Insititute of Science Bangalore.

Panelist - Paul Unbehagen, Alcatel-Lucent
Paul Unbehagen is an active member of the IEEE and IETF.
He has also participated in several IETF WGs to include IS-IS, BGP, L2VPN, and IPVPNs and is currently the author of the IP/SPB IETF draft. He has several patents in communication protocols such as IS-IS and BGP.
Previously Paul has worked in numerous diverse networking environments to include the US Military, Bloomberg, MCI, and Nortel as well as a few startups. Paul thus has 14 years of deployment, operational, network design and architectural experience in live networks ranging from Enterprise to Carrier.

Full Abstract

Speakers
Moderator - Barry Raveendran Greene

Full Abstract

The Research Forum will include:

  • Wide-Area Route Control for Distributed Services

    by Vyatautas Valancius, Georgia Tech

    Many distributed services would benefit from control over the flow of traffic to and from their users, to offer better performance and higher reliability at a reasonable cost. Unfortunately, although today's cloud-computing platforms offer elastic computing and bandwidth resources, they do not give services control over wide-area routing. We propose replacing the data center's border router with a Transit Portal (TP) that gives each service the illusion of direct connectivity to upstream ISPs, without requiring each service to deploy hardware, acquire IP address space, or negotiate contracts with ISPs.

  • Understanding Slow BGP Routing Table Transfers

    by Zied Ben Houidi, France Telecom R&D Orange Labs

    Researchers and network operators often say that BGP table transfers are slow. Despite this common knowledge, the reasons for slow BGP transfers are not well understood. This work explains BGP table transfer delays by combining BGP messages collected at a large VPN provider backbone and controlled experiments with routers of three different vendors as well as a software BGP speaker.

  • Autonomous Traffic Engineering with Self-Configuring Link Weights

    by Srikanth Sundaresan, Georgia Tech

    Network operators use traffic engineering to control the flow of traffic across their networks. Existing TE methods establish static topologies offline, either by setting link weights or by configuring paths a priori. These methods require manual configuration and may not be robust in the face of failures. Some methods also require knowledge about traffic demands and may not be able to handle traffic fluctuations. Even when changes in demand are expected, operators must manually tune network configurations to prepare for them.Because adjusting configurations is difficult to get right, we start from an extreme design point, asking instead whether it is possible to perform traffic engineering online without having to perform any a priori configuration. Our traffic engineering technique, SculpTE, adapts to changing traffic demands by automatically configuring link weights in a stable manner. SculpTE balances load across the network by continually adjusting link weights to expose lightly-loaded paths. We evaluate SculpTE using simulations on realistic ISP network topologies. Our results show that SculpTE achieves excellent load balancing, responsiveness, and stability compared to state-of-the-art TE schemes, without requiring network operators to perform any offline configuration.

Speakers
Moderator - Lixia Zhang, UCLA
Zied Ben Houidi, France Telecom R&D Orange Labs
Zied Ben Houidi is a last year PhD student at UPMC Sorbonne Universités doing his research at Orange Labs, France Telecom R&D, Paris, France. His research interests are in the design and the measurement of data communication networks in general. Prior to his PhD, Zied successfully completed a master degree in Computer Science from UPMC Sorbonne Universités and an engineering degree in Telecommunications from the higher engineering school of communications of Tunis.

Srikanth Sundaresan, Georgia Tech
Srikanth is a third year PhD student at the College of Computing, Georgia Tech, where he works with Professor Nick Feamster. He completed his MS in Computer Engineering at the University of Massachusetts Amherst, where his thesis was on energy aware sensor networks, and his Bachelors in Electrical Engineering at the National Institute of Technology, Jaipur, India. His current research interests include routing and management for future networks, and understanding access network performance for broadband users.

Vyautus Valancius, Georgia Institute of Technology
Vytautas Valancius is a Ph.D candidate at Georgia Institute of Technology, advised by professor Nick Feamster. His research interests include interdomain routing, Internet economics, and network virtualization. Prior to his Ph.D studies, Vytautas obtained M.S. in KTH, Sweden and worked in the networking industry as a systems engineer for 5 years, earning CCIE#14359 certification.

Full Abstract

OpenFlow: An Update
About 20 college campuses are deploying OpenFlow in some part of their network. There appear to be two reasons: 1. It enables their researchers to try out new ideas in a realistic production network, and 2. Because it gives them the ability to add new features to their network.
In this talk I'll give an update on the progress of OpenFlow, why people are using it, and the future standardization plans. In an accompanying talk, Matt Davy (IU and GRNOC) will describe the current and future deployments, and his plans to bring OpenFlow to many campus networks.

OpenFlow Trials and Deployments
OpenFlow is already being tested and deployed at universities and research labs around the world. This presentation will provide an overview of the current OpenFlow trials and deployments globally. Indiana University will share their own experiences from deploying OpenFlow on both enterprise and wide-area research networks as well as their plans for developing and deploying OpenFlow applications to improve how they mange networks.

Speakers
Matt Davy, Indiana University
Matt Davy is the network architect for Indiana University. He provides leadership in the design and planning of research and education networks at the enterprise, regional and national level. He is currently focused on the development and deployment of OpenFlow technologies and the development of the GlobalNOC's internship and training program that provides hands-on networking experience and training to undergraduate and graduate students.

Nick McKeown, Stanford University.
Chris Tracy, ESnet
Chris Tracy is a Network Engineer at the Energy Sciences Network (ESnet) in the Computational Research Division of the Computing Sciences Directorate of the Lawrence Berkeley National Laboratory. ESnet is the high performance networking facility of the US Department of Energy Office of Science. ESnet's mission is to enable those aspects of the DOE Office of Science research mission that depend on high performance networking for success.

Prior to joining ESnet, Chris was a Co-PI on the GENI MANFRED proposal and was one of the systems/optical network engineers for the NSF DRAGON project, a $6.5M research program to deploy "experimental" optical networks utilizing novel technologies and services to provide real and measurable advantage to advanced e-science applications. The program was a collaborative project between the Mid-Atlantic Crossroads, the USC Information Sciences Institute East, George Mason University, and the University of Maryland College Park. Chris has also directly contributed to Internet2's HOPI/DCN project since April 2004.

Chris was also responsible for the deployment and operational management of the DRAGON network - a lambda-switched network which includes over 100 miles of fiber in the Washington DC/Baltimore metro area connecting 7+ POPs with ROADMs, OADMs, L2 switches, routers, etc. He has also helped with strategic planning and providing engineering support for the DRAGON, HOPI, and DCN networks.

Prior to MAX, Chris was a Senior Network Engineer for seven years at a regional ISP in the Pittsburgh, Pennsylvania area. He helped to manage the organization's technical needs, working at every layer from the ISP's physical connectivity up the stack to application/service deployment and management. He has presented at LISA 2002, and was also involved with the Pittsburgh SAGE organization during this time. He has been actively involved with the SCinet planning committee between 2002-2006, primarily working with the IT and WAN groups.

Chris received a Bachelor of Science degree in Computer Engineering from the University of Pittsburgh in 2001, and is currently pursuing a master's degree in telecommunications management at University of Maryland University College (UMUC). He has worked in computing and networking since 1997.

Full Abstract

Alcatel-Lucent

Full Abstract

All first-time NANOG attendees are invited to attend a special continental breakfast where committee members, Merit staff and long-time NANOGers will talk about the organization, the meeting, and how to make best use of both. Please plan to attend and learn more about NANOG and help maximize your first meeting. It will be lively and informative! The breakfast will be moderated by Cat Hoffman of Equinix.Pacnet

Recordings
Full Abstract

Speakers
David Meyer, Program Committee Chair, Cisco/UO
Eric Shepcaro, Telx
Donald Welch, Merit Network

Recordings
Full Abstract

Speakers
Elise Gerich

Recordings
Full Abstract

John will provide an update on the ARIN portion of the joint NANOG/ARIN meeting. As IPv4 and 2-Byte AS run-out dates are quickly approaching, there are a number of issues/policies that are being discussed in ARIN that may be of interest to the community.

Speakers
John Curran, ARIN

Full Abstract

Speakers
Geoff Huston, APNIC

Full Abstract

Telx

Full Abstract

Growing concerns about the use of the current Internet protocol have caused many to consider how it should be changed in order to address problems such as DFZ RIB growth, implementation of certain functionality (such as multi-homing and mobility), as well as dealing with issues of scalability. We take the position that an evolutionary approach is possible for a practical and scaleable solution, giving much of the functionality required, being backwards compatible with the currently deployed architecture, with incremental deployment capability, and which can reduce the current routing state overhead for the core network. By enhancing the way we use naming in the Internet Architecture, it is possible to provide a harmonised approach to the provision of much functionality (such as multi-homing, mobility and others), including specific improvement to the scalability of inter-domain routing state.

Speakers
Saleem Bhatti, University of St Andrews
Saleem Bhatti is a Professor at the School of Computer Science, University of St Andrews, Scotland, UK. His research interests include novel network architectures, mobile and heterogenous networked systems and network security. He is also involved with industry in various consultancy roles in the area of networking technology and systems. He prefers his beer flat, dark, and served at room temperature in a straight glass.

Full Abstract

This presentation aims to bring the low-cost economic 1U/2U top-of-rack device beyond just the datacenter. I intend to outline how you can take some of the low-cost commodity top-of-rack router/switches and transform it into a viable BGP Internet peering router, providing you a dense Nx10GE peering router at an order of magnitude cheaper than the big metal boxes on the market. Through a variety of tricks with MPLS and BGP, an operator can still have roughly the same functionality they have today with a few additional enhancements.

Speakers
Tom Scholl, nLayer
Tom Scholl currently is a Principal Network Engineer at nLayer Communications. He has worked in network engineering roles for the past 14 years and has been an active contributor to the NANOG program with several presentations. He previously was at SBC, which later acquired AT&T and was a key participant on network integrations with Cingular and SBC. While at AT&T, he was main contributor to their architecture of MPLS-TE, BGP and various other network architecture functions. Tom has worked in both engineering and operations roles and understand the needs of each individual group. Tom continues to try to distribute clue and help others when it comes to designing and operating networks.

Full Abstract

This talk will cover: Drivers for LISP, Design, Deployment considerations & results, Operations, IPv6 (LISP6), and Future tasks.

Speakers
Donn Lee, Facebook Inc
Donn Lee is a Sr. Network Engineer at Facebook. His duties include designing networks, evaluating products, optimizing performance, and performing escalation troubleshooting. Previous to Facebook, Donn worked in Google's Network Architecture group for four years and during tremendous growth of Google's backbone, optical, and datacenter networks. While working as a Consulting Systems Engineer at Cisco Systems (CCIE #3262) he worked on large global networks and wrote his book, Enhanced IP Services for Cisco Networks, that is published by Cisco Press. He holds a bachelor's degree in Electrical Engineering from UCLA.

Recordings
Full Abstract

DNS Caching: Running on Zero

Introducing OpenBFDD

In the face of depletion: IPv4 Cost

Speakers
Saleem Bhatti, University of St. Andrews.
Tom Daly, Dynamic Network Services, Inc..
Martin Hannigan, The Iceland Guy.

Tuesday, October 5, 2010
Topic/Presenter
Full Abstract

Follow-up from last BCP BoF

Speakers
Aaron Hughes, Bind

Recordings
Full Abstract

The Day the Root Got Signed
On July 15 at approximately 20:50 UTC, the first signed and validatable root zone file was pushed out to the Internet's root nameserver system. In this presentation I'll talk about (1) VeriSign's infrastructure for generating ZSKs and signing zone files; (2) the steps in transitioning from a DURZ to a validatable root zone; (3) some photos taken at VeriSign on this day; and (4) a graph showing the spike in DNSKEY queries while DNS geeks eagerly awaited the new zone.

Now that the Root is Signed ...
DNSSEC Deployment has always been described a chicken and egg problem. At the publication end, the zone operators have been waiting for an incentive to sign their zones while at the user end, general-purpose DNSSEC validation was seen as inconsequential given the paucity of signed zones. The recent months have noted a marked upswing in the number of signed zones in the Internet. The Root zone is now signed and deployed in production, major TLDs have announced their plans for signing their zones, if they haven't signed already, and the number signed ccTLDs continues to grow. The time is now ripe to start extending the benefits of validation to the end user. This talk will focus on some of the benefits that can be gained from making applications DNSSEC-capable and some of the DNSSEC-capable applications that are available today.

Speakers
Mehmet Akcin, ICANN.
Wes Hardaker, SPARTA/Cobham
Wes Hardaker has been a long time advocate of new secure Internet technologies. He is a strong advocate of secure core-infrastructure technologies such as DNSSEC and SNMPv3. Mr. Hardaker also plays an active role in the IETF where he has authored multiple security related RFCs. He is actively involved with the DNSSEC-Tools project which provides an easy-to-use suite of DNSSEC tools and applications.

Duane Wessels, VeriSign
Following his college education in Physics and Telecommunications, Duane worked on the Squid/IRCache project at UCSD. For many years his company, The Measurement Factory, developed open source testing and measurement tools for HTTP and DNS. Recently he was the Director of the Domain Name System Operations Analysis and Research Center (DNS-OARC). Currently Duane researches DNSSEC deployment for VeriSign.

Full Abstract

Matt Peterson leads network and systems operations at Square, Inc -AS15211. Mr. Peterson's career spans a diverse mixture of non-profit, start-up, enterprise and small business engagements. Matt is well known for his humble, education-without-drama attitude and approachable demeanor. Previously he installed ccTLD DNS root servers around the
global and lead the operations team for one of the most profitable-per-megabit adult video services. The first on-profit Internet Exchange within San Francisco was co-founded by Matt in mid 2006. Mr. Peterson's work has been presented at numerous industry conferences, including: APRICOT, SANOG, BSDCon, Defcon and CCC Camp.

Speakers
Moderator - Matt Peterson

Full Abstract

The pS Performance Toolkit
Debugging network performance end to end can be an intractable problem: availability of tools, knowing where to find them, permissions to use them, and lack of a clear method to view and exchange historical information can thwart attempts to solve problems that span multiple domains.

IRR Additional Route Attributes
The IRR system has traditionally been used by network operators to declare routing policy for their own networks. We propose two additional attributes geoidx and roa-uri which can be added to a route object definition. The geoidx attribute can be used to define a geographic region which that particular prefix is associated with while the roa-uri can be used to reference a roa for that prefix. We have implemented both these new attributes in IRRD as well as example tools that demonstrate how these can be used by various applications.

Speakers
Moderator - Mohit Lad, Thousand Eyes
Mohit Lad is the co-founder of ThousandEyes, a network monitoring and security company. Prior to this, he was responsible for designing and implementing an internal monitoring system for Nokia's global network. He received a Ph.D. in Computer Science from UCLA in 2007 and has been actively involved in NANOG throughout his graduate study at UCLA. During his Ph.D. he designed and implemented the Link-Rank visualization for BGP routing diagnosis. He was also the primary designer of PHAS: Prefix hijack alert system (implemented and maintained by Colorado State University).

Pat Pannuto, Merit Network
Pat Pannuto has been an active University of Michigan
undergraduate student intern at Merit Network Inc for almost 2 years. At Merit he has been involved in the development of various projects such as OpenCALEA, Passive RTT Estimation, as well as RADB. Pat has also interned at Microsoft and Qualcomm and worked on various projects related with
the Linux as well as Windows kernel.

Jason Zurawski, Internet2.

Full Abstract

The NANOG50 Demo Room will feature an IPv6 demonstration with an opportunity to view a live IPv6 dual stack configuration that is currently in production. Technological capabilities of various hardware and software will be featured, allowing you to experience technology, content and services first-hand.

Read more about the Vendor Collaboration Room.A10 Networks

Full Abstract

The Open Policy Hour (OPH) is an informal gathering where folks can talk Internet number resource policy. Would you like to “test drive” a policy idea? Everyone is invited to attend this session and raise ideas and suggestions. You do not need to have a formal presentation in order to participate. Signing up in advance allows us to confirm your turn to present, but is not required.

Part One - Draft Policy Background Briefing: ARIN staff provides a quick overview of the draft policies that will be discussed during the upcoming Public Policy Meeting. The intent here is to increase understanding, not to discuss the pros or cons of the draft policies.

Part Two - Policy Proposal BoF: This is your opportunity to try out a policy idea and get informal feedback. You’ll be given a few minutes to present your idea, as well as a few minutes for questions and answers. You get to say something, and you get feedback. There are no commitments. To sign up to talk, please send an e-mail to policy at arin.net with a brief description of what you want to talk about.

Speakers
Moderator - Einar Bohlin, ARIN

Full Abstract

ARIN would like to welcome you to Atlanta! We are sponsoring a social
for all NANOG and ARIN meeting attendees following the ARIN Open Policy
Hour. We hope you will join us on the Terrace for pizza, a drink, and
a spectacular view of Atlanta's Midtown area. Please come and connect
with friends, old and new, talk to candidates for the ARIN Board,
Advisory Council and NRO Number Council, and recharge before the Public
Policy and Members Meeting gets rolling on Wednesday. Come kick back with us!

Full Abstract

In this talk, we present detailed, quantitative data on 5,000 operator confirmed (via manual classification) network infrastructure attacks over the last year across 37 carriers. We explore the magnitude, frequency and characterize both the attack vectors and victims. Previously, most data on carrier security trends has come from qualitative vendor surveys (e.g. Cisco, Arbor, etc). We believe this talk represents the first large scale study of validated (i.e. "ground-truth") botnet and DDoS attacks. We conclude with estimates of botnet size and efficacy.

Speakers
Craig Labovitz, Arbor Network

Full Abstract

Speakers
Moderator - Nina Bargisen, TDC
Nina has worked at TDC, AS3292, the incumbent in Denmark and one of the leading ISPs in Scandinavia, since 1999, and currently works in the capacity planning group. She is responsible for all for all international interconnects for TDC and runs the IP registry and is part of the technical peering team at TDC. Other responsibilities include network planning, Traffic Engineering, budgeting, network modeling and network design.

Nina has an M.Sc in Mathematics with minor in Computer Science from Århus University Denmark.

Sylvie LaPerrière, Google
Panelist - Per Hansen, Ciena
Per Hansen is currently Senior Director, Submarine Solutions at Ciena, where he is focused on the networking needs of submarine cable operators. Per joined Ciena in September 2008 from ADVA Optical Networking, Prior to joining ADVA in 2004, Per was one of five founding professionals who established Photuris Inc., an early innovator of dynamic optical networking solutions. Earlier, he was with Bell Labs.

Per holds a doctorate in electrical engineering from the Technical University of Denmark and Master of Business Administration from the Wharton School of Business at the University of Pennsylvania.

David Ross, Davis Ross Group
David G. Ross has served the international communications industry since 1970, beginning at Bell Laboratories with development of the technology for undersea communications systems. Over a 30-year corporate career, he served in a variety of managerial positions with Bell Labs, AT&T and Tyco, culminating in the formation of The David Ross Group Inc. (DRG) in 1999.

Mr. Ross and has overseen the growth of DRG into a premier independent consulting and network development firm. Today DRG provides a broad range of technical, commercial, and managerial services to the international telecommunications industry. The firm’s recent activities include over 20 international undersea network developments connecting over 40 countries throughout the world, as well as a broad array of consulting activities from wireless network optimization to telco privatization.

Mr. Ross received the Bachelors and Masters Degrees in Electrical Engineering from the University of Michigan at Ann Arbor. He holds several patents, is a published author, and has been an invited speaker at conferences and seminars worldwide.

Full Abstract

The East African region acquired undersea capacity in mid 2009. Previously, connectivity was via E1 copper lines or in some cases fiber, but to slow VSAT connections. By the end of 2010, three undersea cables will be active. In the research and education area, users made the big leap from shared VSAT to undersea capacity as well. My presentation will show what impact this had and what has needed to be done to adjust to the new bandwidth prices. I will also describe the problems encountered and solutions deployed to counter problems related to the new Internet speeds achieved with undersea capacity.

Speakers
Kevin Chege, KENET
Kevin Chege is the Network Manager for the Kenya Education Network (KENET) which is Kenya's REN. He has been actively involved in setup up of all the infrastructure currently in place in KENET from 2004. He has expertise in network monitoring and management and has contributed content to a book on bandwidth management titled "How to accelerate your Internet" . He has also had an article published by the IETF on Internet connectivity and has been a lead trainer for KENET for a number of years on topics such as bandwidth management and Linux/Unix setup services. He holds a Bsc. in Information Systems and Technology and is currently pursuing a Masters in Information Security from Lulea University of Technology (Sweden) via distance learning.

Full Abstract

ARIN, Global Peering Forum

Full Abstract

In this presentation we'll take a look at what Ethernet developments are taking place in the IEEE, ITU-T, IETF and MEF. We'll start with a quick overview on how the various standards organizations fit together, and then dive deeper into the following new and emerging Ethernet technologies:
- 40/100 GbE
- Energy Efficient Ethernet (EEE)
- Carrier Ethernet
- OAM
- Data Center Bridging (DCB)

Speakers
Greg Hankins, Brocade
Greg Hankins is a Global Solutions Architect at Brocade, specializing in the Ethernet switching and IP/MPLS routing product lines. He works with service providers and Internet exchanges around the world as a technology evangelist, consulting engineer and customer advocate, and is an active
member of the network operator and peering community. Prior to joining Brocade, he held technical marketing and systems engineering positions at Force10 Networks and Riverstone Networks, and network engineering positions
at MindSpring Enterprises and Georgia Tech.

Full Abstract

We present an innovative network monitoring tool that groups under the same "look and feel" interface, access to multiple databases associated with network and client erformance. These include data from our own polling engine Apoll, Spectrum for equipment failure, Nagios for CPU performance, Sflow for traffic flow analysis and environmental for power and cooling.

Speakers
Brian Martin, CERN, Geneva Switzerland
Brian Martin is a senior electronics engineer with over 30 years of experience working at CERN. He has worked on real time data acquisition systems for many experimental collaborations including the Nobel prize winning UA1 experiment. His field of expertise is in high speed busses and interconnects. He has published many technical papers both for CERN and for the several EU funded R&D projects that he has been involved in. He is currently leading the team working on designing commissioning and monitoring the data networks for the ATLAS experiment running at the CERN Large Hadron Collider.

Wednesday, October 6, 2010
Topic/Presenter
Full Abstract

A panel discussion on IPv4 to IPv6 transition and co-existence mechanisms.

Speakers
Moderator - Lee Howard, Time Warner Cable.
Panelist - Yiu Lee, Comcast
Panelist - Jason Weil, Cox Communication
Panelist - Steve Wright, AT&T

Full Abstract

XKL

Full Abstract

A panel discussion on IPv4 to IPv6 transition and Co-Existence experiences.

Speakers
Moderator - Cathy Aronson, Cascadeo Corporation
Cathy began her networking career in 1988 at Merit, Inc where she worked on the NSFNet backbone and also CICNet, a network which interconnected the Big 10 universities. She engineered routing and addressing for BARRNet, the Energy Sciences Network, as well as @Home Network. Cathy was also a member of the technical staff at Packet Design, where she was responsible for operational aspects of their Internet scaling projects. Cathy has served on the ARIN Advisory Council continuously since 1997 and as well as one term on the ASO Address Council.

Panelist - John Brzozowski, Comcast.
Panelist - Tom Coffeen, Limelight Networks
Panelist - Lorenzo Colitti, Google
Panelist - Michael Sinatra, UC Berkeley
Panelist - Jason Weil, Cox Cable

Full Abstract

Speakers
David Meyer, Program Committee Chair, Cisco/UO.
Andy Rosenzweig, Merit Network