Sunday, June 9, 2019
Topic/Presenter
Full Abstract

Oracle

Recordings
Full Abstract

Oracle

Full Abstract

Equinix

Full Abstract

Amazon Web Services

Full Abstract

In 2009, we presented results from a large scale, multi-year study of global Internet traffic across 110 providers and 200 Exabytes of commercial traffic. Our original study found significant changes in inter-AS traffic patterns, an increasing consolidation of "HyperGiant" traffic volumes and an evolution of provider peering strategies[. In this talk, we revisit our work from a decade ago with an in-depth look at the changes to traffic patterns, content delivery and Internet peering across more than 50 providers around the world. We show how cloud, IoT and significant changes to interconnection strategies are impacting networks in North America, Europe and Asia.

Speakers
Craig Labovitz, Nokia

Speakers
  • 3 Craig Labovitz
Damian Menscher - Google
Damian Menscher - Google
Full Abstract

We review the growth of DDoS attacks over the past decade, with a focus on how they're generated. We then show practical methods for tracing the origin of amplification attacks, and ultimately filtering spoofed packets near the source.

Speakers
Damian Menscher, Google
Damian Menscher is responsible for DDoS defense at Google, where he has studied hundreds of attacks over the past decade. He uses his experience to design automated defenses for common attacks. Prior to joining Google, Damian completed a PhD in computational particle physics.

Damian Menscher: Damian Menscher is responsible for DDoS defense at Google, where he has studied hundreds of attacks over more than a decade. He uses his experience to design automated defenses for common attacks. Damian has a Ph.D. in physics from UIUC.
Damian Menscher: Damian Menscher is responsible for DDoS defense at Google, where he has studied hundreds of attacks over more than a decade. He uses his experience to design automated defenses for common attacks. Damian has a Ph.D. in physics from UIUC.
Speakers
  • 3 Damian Menscher - Google
  • Speaker Damian Menscher - Google
Full Abstract

As network engineers, it’s almost impossible to avoid the CI/CD/NetDevOps hype-cycle. In this talk we explore how much of this is a myth, versus the ground reality network engineers experience, managing production networks. In our experience, while CI/CD for most vendors can demo very well, they do not hold ground for production networks. The interdependencies between physical hardware and the Network OS (NOS) make it difficult to conduct realistic tests on a purely virtual form factor. In this talk we propose a realistic approach using GitOps and Infrastructure as Code. We will also discuss the limitations of virtualized network OS and provide recommendations on utilizing virtualized platforms for limited testing and configuration linting

Speakers
Ajay Chenampara
Ajay is a network industry veteran with over 18 years in the space. He is a Domain Architect with Red Hat, focused on helping customers achieve their business outcomes, using Ansible for automating their networks. Previously he was the global datacenter architect for a top 10 Fortune 500 enterprise, leading the network automation efforts there and subsequently worked for a community focused network automation startup, helping network engineers adopt DevOps tools and methodologies across the globe.

Gerald Dykeman
Ansible Domain Architect for North America Public Sector with a focus on Network Automation and Programmability. Previously at Cisco Systems working on a range of solutions from Enterprise to Cloud technologies before moving to Red Hat. CCIE #36354

Speakers
  • 3 Ajay Chenampara
  • Gerald Dykeman
Alan Clegg - ISC
Full Abstract

This presentation will talk you through some of the realities concerning migrating from one DHCP server to another. Using real-world examples, Alan will address the issues that arise when making changes to core network infrastructure. The talk goes on to cover how to take an existing, working configuration for ISC DHCP, and use it to bootstrap a configuration for a Kea DHCP server, using ISC's Migration Assistant tool. Translating your existing configuration using this tool will also help you verify whether Kea supports all the features of ISC DHCP you are currently using.

Speakers
Alan Clegg, ISC

Speakers
  • 3 Alan Clegg - ISC
Monday, June 10, 2019
Topic/Presenter
Full Abstract

QTS Richmond NAP

Recordings
Full Abstract

There’s been a lot of talk about scaling and operational challenges of running datacenter CLOS networks, with some people saying that there isn’t enough information about exactly how to build them well, and others saying that it’s too hard, and that we need new protocols created to help with the problem. This talk will decloak the CLOS design that Yahoo used to underpin its datacenter networks starting 2012, describe the architectural tradeoffs of different CLOS designs, talk about specifics of the design selected, as well as describe lessons learned in building, scaling, and operating those networks.

Speakers
Igor Gashinsky, Verizon Media / Yahoo!
Igor is the Chief Network Architect at Verizon Media / Yahoo!, where his responsibilities range from overall network design, including highly resilient datacenter and backbone switching and routing architecture, peering strategy, MPLS design, and L4-7 loadbalancing, to distributed and scalable content delivery methodologies, DNS architecture, and R&D into future technologies.

Yihua He
Yihua He is a principal network architect in Yahoo. He holds a PhD degree in computer science and has numerous publications in highly cited computer network conferences and journals. In the past four years, he has focused on researching, designing, prototyping and implementing Yahoo's next generation data center networks. In this presentation, he would like to share the experience of a state-machine based configuration management system that automates daily operations of a data center network end-to-end.

Full Abstract

Sparkle

Full Abstract

Iron Mountain Data Centers

L Sean Kennedy
Mike Hagan - EdgeMicro
Benson Schliesser
Edward McNair - NANOG
Recordings
Full Abstract

Kickoff of NANOG 76 in Washington D.C. with PC Chair, NANOG Representative and a word from our Host Sponsor.

Speakers
L Sean Kennedy
L Sean Kennedy is an active member of the Internet Engineering community and Chairperson of the NANOG Board of Directors.

Mike Hagan, CEO/Co-Founder, EdgeMicro
Benson Schliesser, Volta Networks

Valerie Wittkop: Valerie has worked with the North American Network Operators’ Group (NANOG) community since 2006 in various roles including support of the Program, Communication, and Development Committees, as well as Annual Elections. In 2015 she became the Program Director for NANOG and continues to work in that roll today. With a demonstrated history of working in the Internet industry, she is a business development professional with skills in non-profit management, fundraising, networking, sales, and start-ups. Valerie holds a BS focused in Business Management from University of Phoenix.
Edward McNair: Edward McNair is the Executive Director of the North American Network Operators Group (NANOG). He is also the co-founder of Kaskadian, an agency that provides branding, marketing and sales support for startups and new businesses. Prior to Kaskadian, Edward served as Chief Executive Officer for Verilan, an IT company that delivered just-in-time, enterprise-quality networks. Previously, he was Vice President of Internet Marketing for R2C, a leading direct marketing agency, and was Creative Director for the WiMAX Forum, a global Internet and telecom consortium. In the computer industry, Edward has developed corporate training solutions for Nike, Adidas, Columbia Sportswear, Kaiser Permanente, and FEI, among others. In addition, he has delivered professional services to NANOG, the Institute of Electrical and Electronics Engineers (IEEE), the Internet Corporation for Assigned Names and Numbers (ICANN), Facebook, Intel® and Mentor Graphics. Edward also developed the first web design program at the Pacific Northwest College of Art where he taught web and graphic design and interactive media courses for more than a dozen years. In his free time, Edward is involved in community theatre aimed at supporting local charities. His most recent production was playing the lead in the musical "Oklahoma!”
Speakers
Full Abstract

In five minutes on Monday morning John Curran, President and CEO of ARIN would like to invite members of the NANOG community to share their experience in an IPv6 case study and join ARIN in our quest to move the needle further toward IPv6 adoption. To date, ARIN has collected over 20 cases studies from a wide range of organizations, including a Tier 1 global provider who reported that in some ways IPv6 is easier than IPv4, another organization explaining why they chose to develop an IPv6-only internal network, tips on how to talk to vendors, and why and how many academic institutions took an early adopter mind-set. ARIN will have a staff member at its Registration Services Help Desk during NANOG devoted to talking with organizations interested in sharing their story, and they can even get started onsite. We have a series of questions to ask, can record the conversation and provide the organization with a first draft to refine for publication. It’s a relatively low commitment of time to share your experiences, and your insights may help expedite someone else’s journey. By working together, ARIN and NANOG can provide a helpful resource. We are looking for organizations willing to promote their hard work to deploy IPv6 and help the community at the same time. ARIN's IPv6 deployment case studies are available at: https://teamarin.net/get6/ipv6-case-studies/

Speakers
John Curran, American Registry for Internet Numbers (ARIN)
John Curran is the President and CEO of ARIN. John is a recognized expert on Internet and telecommunications matters and has been the Chief Technology Officer of three successful Internet companies: BBN/GTE-Internetworking, XO Communications, and ServerVault Corporation. He has also been an active participant in the Internet Engineering Task Force (IETF), having co-chaired the IETF Operations and Network Management Area and served as a member of the IPng (IPv6) Directorate. John is a founding member of the ARIN Board of Trustees, having served since August 1997. He was re-elected in 2000, 2003, and again in October 2006, and served as Chairman of the Board of Trustees from inception through early 2009. Since 2009, John has served as ARIN's President and CEO of ARIN, which is a non-elected, full-voting member of the Board of Trustees.

Speakers
  • 3 John Curran - ARIN
Full Abstract

Do you re-assign IP address blocks to customers? Ever use WHOIS? Operate a network on more than one continent? Do you have feelings about ARIN’s IPv4 waiting list? All of the above?! ARIN policies impact the network operator community. This talk will cover the policy discussion that is currently underway in the ARIN region with respect to the above named issues and more. It is also an opportunity for interested network operators to provide feedback without investing the resources to attend a full ARIN meeting. We want to know whether the concepts raised in the current policy proposals are sound, understand how proposals would affect operators, and discern how to best minimize that impact while maximizing value to the community. Please note that this session is not intended for debate of the specific language of the policy proposals.

Speakers
Alyssa Moore, CIRA

Speakers
  • 3 Alyssa Moore - CIRA
Recordings
Full Abstract

Many businesses would benefit from IPv6. Is there a way to calculate how much? Can we find a way to solve the prisoner's dilemma, where deploying IPv6 is only helpful if everyone else deploys IPv6.

Speakers
Lee Howard, Retevia

Full Abstract

There’s been a lot of talk about scaling and operational challenges of running datacenter CLOS networks, with some people saying that there isn’t enough information about exactly how to build them well, and others saying that it’s too hard, and that we need new protocols created to help with the problem. This talk will decloak the CLOS design that Yahoo used to underpin its datacenter networks starting 2012, describe the architectural tradeoffs of different CLOS designs, talk about specifics of the design selected, as well as describe lessons learned in building, scaling, and operating those networks.

Speakers
  • 3 Igor Gashinsky
  • Yihua He
Susan Forney - Hurricane Electric
Full Abstract

Speakers
Susan Forney, Hurricane Electric
Netflix

Susan Forney: Susan Forney is a network engineer at Hurricane Electric Internet Services, which operates the largest IPv6 backbone in the world in terms of number of connected networks. Before joining Hurricane Electric, she was the principal network engineer at the Markley Group, where she developed and operated the Boston Internet Exchange. She was a principal network engineer and architect at at the Microsoft Corporation for 15 years, and also worked as a network engineer at Starbucks and the Boeing Company. She is a member of the Board of Directors of the North American Network Operators Group (NANOG).
Speakers
  • 3 Susan Forney - Hurricane Electric
Full Abstract

Kentik

Recordings
Full Abstract

Speakers
Syed Ahmed

Speakers
  • 3 Syed Ahmed
Paul Ebersman - Neustar
Paul Ebersman - Neustar
Full Abstract

DNS hijacking has become a mainstream technique for subverting websites, stealing login credentials, fraud and abuse. The DNS is a complex system with many attack surfaces. Defending the integrity of your DNS is more critical than ever but how? DNSSEC is a useful layer of a defense in depth of your DNS but it isn't the only layer or technique. Learn in detail what DNSSEC can defend against and what else you need to do.

Speakers
Paul Ebersman, Neustar
Paul Ebersman has been involved with NANOG to varying degrees since the late 90s and has been working with TCP/IP networks since the mid 80s. Paul Ebersman works for Neustar as a DNS architect and as a technical resource, both internally and to the internet community. He first worked on the internet for the Air Force in 1984. He was employee number ten at UUNET and helped build AlterNET and the modem network used by MSN, AOL and Earthlink. He has maintained his roots in the internet and the open source community, with heavy involvement in organizations like NANOG, IETF, ICANN/SSAC and DNSOARC.

Paul Ebersman: Paul Ebersman has been involved with NANOG to varying degrees since the late 90s and has been working with TCP/IP networks since the mid 80s. Paul Ebersman works for Neustar as a DNS architect and as a technical resource, both internally and to the internet community. He first worked on the internet for the Air Force in 1984. He was employee number ten at UUNET and helped build AlterNET and the modem network used by MSN, AOL and Earthlink. He has maintained his roots in the internet and the open source community, with heavy involvement in organizations like NANOG, IETF, ICANN/SSAC and DNSOARC.
Paul Ebersman: Paul Ebersman has been involved with NANOG to varying degrees since the late 90s and has been working with TCP/IP networks since the mid 80s. Paul Ebersman works for Neustar as a DNS architect and as a technical resource, both internally and to the internet community. He first worked on the internet for the Air Force in 1984. He was employee number ten at UUNET and helped build AlterNET and the modem network used by MSN, AOL and Earthlink. He has maintained his roots in the internet and the open source community, with heavy involvement in organizations like NANOG, IETF, ICANN/SSAC and DNSOARC.
Speakers
  • 3 Paul Ebersman - Neustar
  • Speaker Paul Ebersman - Neustar
Justin Kilpatrick
Full Abstract

Althea (althea.org) which is an open source 'distributed isp' or 'isp in a box', as such we perform traffic shaping, routing, encryption, and billing entirely on home wifi routers re-flashed with a open source OpenWRT image of our own design. The goal is to eliminate the need for any sort of configuration or network engineering from the entire process of being an ISP, everything is as easy as plugging one device into the next. Or at least it should be. This talk will go into detail about the open source technologies used and problems encountered in creating an modern and reliable ISP where no one is in charge. Main topics * Babeld, a link detecting auto failover 'mesh' protocol that's reliable, stable, and easy to use. * Wireguard, an ultra low overhead VPN protocol that provides incredible security at an overhead acceptable for even embedded devices * Bufferbloat.net (http://bufferbloat.net/) (Specifically fq_codel and Cake) lag spikes are obsolete. There's no reason to be afraid of queue management anymore. * Tales from the field, what operating and using a system like this is like.

Speakers
Justin Kilpatrick, althea.org
CTO and Lead developer for Althea. Free software fanatic, Linux expert, former Red Hatter

Speakers
  • 3 Justin Kilpatrick
Full Abstract

Artificial Intelligence and the related field of Machine Learning is being increasingly used in applications to learn from data in diverse domains. This has been enabled in recent years by increased data availability and improved ML/AI techniques. This talk will provide an overview of Machine Learning/Artificial Intelligence terminology and will look at different Machine Learning techniques that can be used to understand network behavior. It will specifically cover the Machine Learning techniques of classification, anomaly detection and prediction using real world examples. It will also look at the challenges in learning from network data and the benefits and limitations of Machine Learning applied to networking.

Speakers
Srividya Iyer
Srividya Iyer has over 15 years of experience in the networking industry. She has worked at Motorola Mobility in the areas of Metro Ethernet and Seamless Mobility. Prior to that she was at Telcordia Technologies working on Fault and performance management systems. Currently, she is involved in the Research and Development of Machine Learning solutions for networking and has received a National Science Foundation grant for researching Self adaptive Machine Learning models for event correlation.

Speakers
  • 3 Srividya Iyer
Full Abstract

Flooding, hurricanes, tornadoes, fires and other severe weather/environmental events can dramatically impact our ability to deliver IP based communications to end users, which can include emergency services. While the primary use of BGP is interconnection of ISP's, enterprise and government network users can also benefit from the ability to deploy BGP on the edge for multi-homing and redundancy purposes. A many-pathed network becomes significantly more robust for every participant. Direct carrier participation in paid, filtered peering arrangements can contribute to the solution, as can IXP based encrypted tunnel endpoints providing BGP sessions to the edge. With the increase in severe weather events effecting more and larger service areas, additional effort is required to ensure that our networks don't fail as a result of a more hostile, changing climate, just as packet switching ensured continued communications between key facilities in the event of nuclear attack.

Speakers
Scott Johnson, SolarNetOne, Inc.

Alexander Azimov - Yandex
Alexander Azimov - Yandex
Full Abstract

The BGP routing protocol was designed to control traffic at interdomain routing level, but Its scalability and extensibility made it popular in other environments: FlowSpec, VPN, SD-WAN, and other technologies relies on underlying BGP transport. Unfortunately, this diversity of applications haven’t changed the BGP protocol itself – the protocol communications are built on trust, trust in good intentions of all parties, and the trust doesn’t scale that much. In recent years there was a growing hacker activity in BGP with confirmed redirection to the fishing sites, lost of credentials, etc. The community has very limited technical opportunity to fight this threat: most of the filtering measurements are limited to detection of mistakes, others are hardly deployable. During this report, I will provide an overview of previous security mechanisms that were designed to detect malicious routes in BGP and present a novel approach called Autonomous System Provider Authorization (ASPA) that can fill the gap and significantly limit opportunities for attackers.

Speakers
Alexander Azimov, Yandex
Alexander Azimov is network expert at Yandex, where he focuses on the network monitoring, routing security and transport layer architecture. He is a frequent presenter at network operator events such as NANOG, RIPE, ENOG & APRICOT for over 5 years. Alexander is active in the IETF where he co-authoring several Internet-Drafts.

Speakers
  • 3 Alexander Azimov - Yandex
  • Speaker Alexander Azimov - Yandex
Full Abstract

Abstract: Networking and software have had a challenged relationship. While network operators could benefit from Open Source Software (OSS) going forward, OSS hasn’t traditionally met operator needs, and operators may not have the time or resources to contribute to it. This talk will introduce an information-gathering survey of network operators to gather information about: + Scoping – would there be interest in reference implementations for new and existing open standards? Horizontal implementation of networking suites to facilitate new development (e.g., IoT, whitebox routing)? Opportunistic development of specific network management and monitoring tools? + Support and uptake – what are the primary factors in ensuring uptake of the resulting software? (E.g., involvement in development, having paid support plans available, etc). This lightning talk will introduce the survey, solicit input to help shape the questions to achieve a useful result, and help shape future OSS development for network operations. The survey itself is expected to run in August 2019.

Speakers
Leslie Daigle, ThinkingCat Enterprises

Leslie Daigle: Leslie Daigle has been working at the intersection of technology, business/economics and policy to drive effective change for more than twenty years. Leslie is currently the Chief Technical Officer and Director of the Internet Integrity Program at the Global Cyber Alliance (GCA), furthering GCA’s development and deployment of global solutions that contribute to eradicating cyber risk. Leslie is the Principal at ThinkingCat Enterprises, as well as co-founder and co-host of the TechSequences podcast, which explores the many facets of Internet technology, along with its intended (and sometimes unintended) consequences.
Full Abstract

With IPv6 adoption increasing on the content side, it's becoming more obvious that not all IPv6 implmentations are created equal - or are on par with IPv4. In this particular case, it turns out that a lot of CPEs don't have hardware support for a common workaround to implement IPv6 on networks that don't natively support it. For internet connections over 100Mbps, which are becoming more common now, this causes hot CPEs and poor throughput of IPv6.

Speakers
Remco van Mook

Speakers
  • 3 Remco van Mook
Full Abstract

QTS Richmond NAP

Tuesday, June 11, 2019
Topic/Presenter
Full Abstract

Charter Communications

Krassimir Tzvetanov - Purdue University
Krassimir Tzvetanov - Purdue University
Full Abstract

In this tutorial the attendees will go over the basics of Denial of Service. It starts with coverage of the different parts of the stack that can be attacked and transitions into a discussion about the most prevalent types of DDoS: reflection attacks, SYN flood, Sloworis, etc. While it covers different attack types, it supplements the attack descriptions with detailed technical explanation of the specific operating system components like sockets, buffers, etc. The class is interlaced with a number of exercises allowing the attendees to manually configure different mitigations. This tutorial focuses on the technologies and not on particular vendor implementation. The duration can be 60-90 minutes.

Speakers
Krassimir Tzvetanov, Fastly, Inc.
Krassimir Tzvetanov is a security engineer at Fastly, a high performance CDN designed to accelerate content delivery as well as serve as a shield against DDoS attacks. In the past he worked for hardware vendors like Cisco and A10 focusing on threat research, DDoS mitigation features, product security and best security software development practices. Before joining Cisco, Krassimir was Dedicated Paranoid (security) at Yahoo!, Inc. where he focused on designing and securing the edge infrastructure of the production network. Part of his duties included dealing with DDoS and abuse. Before Yahoo! Krassimir worked at Google, Inc. as an SRE for two missing critical systems, the ads database supporting all incoming revenue from ads and the global authentication system which served all of the company applications. Krassimir holds Bachelors in Electrical Engineering (Communications) and Masters in Digital Forensics and Investigations.

Speakers
  • 3 Krassimir Tzvetanov - Purdue University
  • Speaker Krassimir Tzvetanov - Purdue University
Full Abstract

Salt is currently one of the largely used frameworks in automating network operations. Even though it has been firstly released over 8 years ago, the network community started to adopt Salt only after 2016 when the initial official support for automating network gear has been added into the official releases. NAPALM has been one of the first libraries fully integrated into the framework, and others followed later, now being able to automate a variety of platforms, including Juniper, Arista, Cisco IOS / IOS-XR / NX-OS, Palo Alto, Cumulus, as well as many other NAPALM plugins available through the parent library - see https://github.com/napalm-automation-community/. Salt has been designed from the beginning as an event-driven pipeline, easily customizable and extensible in the user's environment. Thanks to Salt's maturity and the features added in the latest releases, it is now easier than ever to write yourself features for network automation by simply executing direct API calls, in just a few lines of Python code. In this tutorial we well explore the available features in the latest Salt release 2019.2.0, and, step-by-step, we'll implement together various features for network operations, from the simplest examples to more complex ones. To follow my steps together with me, you'll only need to have access to a machine where you can run Docker (see https://docs.docker.com/install/ to prepare in advance).

Speakers
Mircea Ulinic, DigitalOcean

Speakers
  • 3 Mircea Ulinic
Full Abstract

Adva Optical Networking

Full Abstract

Opengear

Chris Pennisi
NIgel Bayliff
Erick Contag
Brian Lavalle
Tim Stronge - TeleGeography
Full Abstract

Telegeography’s Tim Stronge will present an overview of the recent and upcoming subsea cables landing into North America. Tim will then be joined by an expert panel to discuss the technical and commercial impacts of these new systems. Topics to include the demand driving these new builds, the technology underpinning the massive capacity gains per system, the growing diversification of North American landing points, and the shift/implications of subsea cable ownership towards content providers.

Speakers
Moderator - Tim Stronge, TeleGeography
Tim Stronge is Vice President of Research at TeleGeography. His areas of expertise include international voice traffic, terrestrial and submarine cable systems, and international bandwidth markets. Since joining TeleGeography in 1996, Tim has served as a principal analyst in most areas of research, including network infrastructure, bandwidth demand modeling, cross-border traffic flows, and telecom services pricing. He holds a Master's degree in International Economics from John Hopkins University and a B.A. from the College of William and Mary.

panelist - Nigel Bayliff
panelist - Erick Contag
panelist - Brian Lavalle

Speakers
  • 3 Chris Pennisi
  • Moderator NIgel Bayliff
  • Erick Contag
  • Brian Lavalle
  • Panelist Tim Stronge - TeleGeography
Full Abstract

This talk will cover the core network requirements for 5G transport. I will cover the 5G architecture on high level pointing out that 5G services are all "softwarized" i.e. cloud native with NFV and SDN enabling flexibility in service deployments. This means that you still need to have a core network (wired) in place to get to your virtualized services like vRAN etc residing in a DC somewhere. Also MEC will be discussed with "micro DCs" closer to the user. This talk will walk the audience through things like network slicing, TSN, automation in relation to the new 5G services like Massive IoT, Mission Critical Services, /FixedMobile Broadband etc. Link to this talk given at an other event: https://www.youtube.com/watch?v=f9ZpMorsXhU

Speakers
Mikael Holmberg
Mikael Holmberg is an experienced networking professional that has over 25+ years experience in Telecom and Networking industry, he is acting as a subject matter expert in networking architectures and technologies.

Full Abstract

Public Speaking Forum provides a positive and supportive environment for participants to practice public speaking skills. All levels are welcome. The first hour starts with introduction, followed by individual presentations, and ended with table topics. The remaining 30 mins are networking time. If you are interested in improving public speaking skills, helping others improve, or networking, this activity is for you. Feel free to use this opportunity to dry run your next presentations. Six 4-min speaking slots are available. Sign up is required if you want a speaking slot. First come, first served. You can sign up via the following URL: https://archive.nanog.org/meetings/nanog76/psf

Speakers
Christina Chu, NTT America
Christina serves as the Director of IP Strategic Planning in NTT America, Inc. She has over 20 years of experience in the Internet industry and is passionate in building communities. She has served four years in the NANOG Program Committee. She currently serves the Global Peering Forum board. Having participated in Toastmasters provides her firsthand experience the benefit of constant practice in improving one’s public speaking skills. She would love to help NANOG put together this public speaking program to nurture the community’s interest in presenting on stage.

Full Abstract

This presentation would address the current problems servicing rural and tribal areas, the reasons for, and new initiatives to fix these problems. Rural and tribal broadband has been a problem since the first long-haul routes were built. The model of economic incentive at the time was not conducive to landing fiber in sparsely populated areas or where right-of-way owners caused delays or impasses. The divide between the haves (broadband, cell) and have-nots has now caused a socio-economic and cultural catastrophe that few could have imagined. There is a changing tide that is now allowing these routes to be built. The primary driver is the rise of the hyperscale content provider and their need for diversity and capacity that is allowing new routes to be built. We can capitalize from these routes by connecting the formerly bypassed areas that are along or near the new running lines. There are significant changes in tribal politics which are also providing access to trusted partners where none existing prior. Add the government economic incentives and the time is right to finally fix this inequity that we take for granted in heavily connected areas. This presentation will not deep dive into the technical details of fiber. If specifics are desired, construction information can be presented, but that would most likely be better served with company specific information outside of this presentation. This presentation would discussing the changing climate and what it means to these affected areas, as well as present the benefit to the content providers, carriers, partners, and end users. The presentation would be given by Matt Rantanen, Southern California Tribal Chairmen’s Association, Tribal Digital Village Network/Initiative, Chairman of the Board for Native Public Media, FCC Native Nations Broadband Task Force. Matt is of Cree ancestry and is involved with all aspects of empowering tribal lands through technology. He's also 6-'6", fully tattooed, and hangs out with Slayer.

Speakers
Matt Rantanen, Arcadian Infracom / SCTCA
Matthew Rantanen, Dir of Technology: Southern California Tribal Chairmen's Association & Director: Tribal Digital Village Network & Lead: Partnering & BizDev at ARCADIANINFRA.COM, a fiber infrastructure company, bringing fiber through rural USA, & Indian Country. Of Cree, Finnish, & Norwegian decent, a "cyber warrior for community networking." helping 20 SCTCA Tribes in technology vision. 2 Terms, FCC Native Nations Broadband Task Force (Genachowski & Wheeler). Appointed to FCC: Communications Security, Reliability and Interoperability Council 4. Former 2-term Chairman at Native Public Media, Vice-Chair & Treasurer. Co-Chair:Tech & Telecom Subcommittee: National Congress of American Indians. Drafting policy with Tribes for use with the Federal Government. Matthew speaks to support unserved/under-served communities, with local, state & Federal government departments. Speaking internationally at ICANN, Internet Governance Forum(UN) Testified at FCC hearings, GAO and the California Public Utilities commission.

Christina Chu - NTT
Full Abstract

Public Speaking Forum provides a positive and supportive environment for participants to practice public speaking skills. All levels are welcome. The first hour starts with introduction, followed by individual presentations, and ended with table topics. The remaining 30 mins are networking time. If you are interested in improving public speaking skills, helping others improve, or networking, this activity is for you. Feel free to use this opportunity to dry run your next presentations. Six 4-min speaking slots are available. Sign up is required if you want a speaking slot. First come, first served. You can sign up via the following URL:
https://archive.nanog.org/meetings/nanog76/psf

Speakers
  • 3 Christina Chu - NTT
Full Abstract

This talk will cover the core network requirements for 5G transport. I will cover the 5G architecture on high level pointing out that 5G services are all "softwarized" i.e. cloud native with NFV and SDN enabling flexibility in service deployments. This means that you still need to have a core network (wired) in place to get to your virtualized services like vRAN etc residing in a DC somewhere. Also MEC will be discussed with "micro DCs" closer to the user. This talk will walk the audience through things like network slicing, TSN, automation in relation to the new 5G services like Massive IoT, Mission Critical Services, /FixedMobile Broadband etc.

Link to this talk given at an other event: https://www.youtube.com/watch?v=f9ZpMorsXhU

Speakers
  • 3 Mikael Holmberg
Full Abstract

Flooding, hurricanes, tornadoes, fires and other severe weather/environmental events can dramatically impact our ability to deliver IP based communications to end users, which can include emergency services. While the primary use of BGP is interconnection of ISP's, enterprise and government network users can also benefit from the ability to deploy BGP on the edge for multi-homing and redundancy purposes. A many-pathed network becomes significantly more robust for every participant. Direct carrier participation in paid, filtered peering arrangements can contribute to the solution, as can IXP based encrypted tunnel endpoints providing BGP sessions to the edge. With the increase in severe weather events effecting more and larger service areas, additional effort is required to ensure that our networks don't fail as a result of a more hostile, changing climate, just as packet switching ensured continued communications between key facilities in the event of nuclear attack.

Speakers
  • 3 Scott Johnson - Spacely Packets, LLC
Full Abstract

As any reader of the NANOG mailing list knows, the push to implement the Resource Public Key Infrastructure (RPKI) has gained momentum over the past year. AT&T has begun dropping invalid routes, NTT uses RPKI data to generate filters, and Cloudflare has developed new validation software. At the same time as these technical efforts have been undertaken, the American Registry of Internet Numbers (ARIN) has also undertaken a review of its legal arrangements regarding RPKI. This lightning talk will provide brief background to the uninitiated and then update attendees on discussions held regarding RPKI at the recent ARIN meeting in Barbados. It will describe the key issues under discussion and the potential approaches ARIN might take to resolving them. It will also highlight recommendations about the best potential approaches to resolving these issues, based on research conducted with Christopher Yoo and funded by National Science Foundation Award #1748362.

Speakers
David Wishnick, University of Pennsylvania
David Wishnick is an Academic Fellow at the University of Pennsylvania Law School’s Center for Technology, Innovation, and Competition. David’s scholarship focuses on the interactions between law and technology in shaping commercial transactions and business forms. Before joining Penn, David practiced at Jenner & Block LLP in Washington, DC, where he advised clients in the finance and communications industries. Prior to joining Jenner, David clerked for Judge Guido Calabresi of the U.S. Court of Appeals for the Second Circuit and Judge Thomas B. Griffith of the U.S. Court of Appeals for the D.C. Circuit. David holds a J.D. from Yale Law School and an A.B. magna cum laude from Brown University. He was born and raised in Chicago.

Speakers
  • 3 David Wishnick
Full Abstract

I presented on DNS Flag Day 2019 at NANOG75. The 2019 Flag Day was a success with very little disruption to production DNS services. This success has emboldened the DNS community to propose a second Flag Day to be held in 2020 (date TBD). The DNS community did receive feedback from many operators that enough notice was not given for the 2019 event. This talk is an attempt to not repeat that same issue. I'll discuss what is being remedied with the 2020 Flag Day: problems with IP fragmentation of DNS packets IP fragmentation is a problem on the Internet today, especially when it comes to large DNS messages. These issues can be fixed by honoring an EDNS buffer size that will not cause fragmentation and by allowing DNS to switch from UDP to TCP when larger buffer sizes are not enough.

Speakers
Eddy Winstead, Internet Systems Consortium
Eddy has over 20 years of DNS, DHCP and sysadmin experience. He was a systems analyst and hostmaster for the North Carolina Research and Education Network (NCREN) for over a decade. At ISC, Eddy has delivered DNS + DNSSEC consulting, configuration audits and technical training.

Speakers
  • 3 Eddy Winstead
Full Abstract

Understanding analytical engines is essential to any analytical product regardless of market. Network analytic tools are no exception. Analytical engines greatly influence use case definitions. Engine defines use case types, and at the same time, limits use case types. Network analytics is plagued by the same restrictions – all require an analytics engine: - Peering analysis - DDOS detection - Subscriber Analytics - Routing analytics This presentation will look at the general available engines and their benefits and detriments. It will also help you think of your current deployments and future analytics purchases to help understand if the product engine is up to the task.

Speakers
Duncan Pauly, CTO, Edge Intelligence
Duncan Pauly is a serial entrepreneur and holder of numerous patents related to data management and database design. He possesses over two decades of senior technical experience with roles involving intellectual property development, high performance software architecture, product strategy and technical governance. He founded numerous companies including CopperEye for high-performance data management, Zenulta for high-performance event correlation and Xi Systems for telecommunication customer care and billing systems.

Speakers
  • 3 Duncan Pauly
Full Abstract

When AMS-IX faced the need for a data center network refresh and upgrade, we evaluated traditional vendors and newer open, bare-metal switching options and found there were numerous reasons to go open. This presentation discusses our use case (a multi-site DC management network), the factors that drove our decision to deploy an open network, and lessons learned from our deployment and over a year of operational experience.

Speakers
Bart Myszkowski

Speakers
  • 3 Bart Myszkowski
Full Abstract

QTS Richmond NAP

Edward McNair - NANOG
Full Abstract

https://youtu.be/bqGHMiSA1Wc

Speakers
Edward McNair, NANOG

Edward McNair: Edward McNair is the Executive Director of the North American Network Operators Group (NANOG). He is also the co-founder of Kaskadian, an agency that provides branding, marketing and sales support for startups and new businesses. Prior to Kaskadian, Edward served as Chief Executive Officer for Verilan, an IT company that delivered just-in-time, enterprise-quality networks. Previously, he was Vice President of Internet Marketing for R2C, a leading direct marketing agency, and was Creative Director for the WiMAX Forum, a global Internet and telecom consortium. In the computer industry, Edward has developed corporate training solutions for Nike, Adidas, Columbia Sportswear, Kaiser Permanente, and FEI, among others. In addition, he has delivered professional services to NANOG, the Institute of Electrical and Electronics Engineers (IEEE), the Internet Corporation for Assigned Names and Numbers (ICANN), Facebook, Intel® and Mentor Graphics. Edward also developed the first web design program at the Pacific Northwest College of Art where he taught web and graphic design and interactive media courses for more than a dozen years. In his free time, Edward is involved in community theatre aimed at supporting local charities. His most recent production was playing the lead in the musical "Oklahoma!”
Speakers
  • 3 Edward McNair - NANOG
Wednesday, June 12, 2019
Topic/Presenter
Recordings
Full Abstract

A chance to update the Community on NANOG events and plans, open forum discussions, Q&A.

Speakers
Edward McNair, NANOG

Edward McNair: Edward McNair is the Executive Director of the North American Network Operators Group (NANOG). He is also the co-founder of Kaskadian, an agency that provides branding, marketing and sales support for startups and new businesses. Prior to Kaskadian, Edward served as Chief Executive Officer for Verilan, an IT company that delivered just-in-time, enterprise-quality networks. Previously, he was Vice President of Internet Marketing for R2C, a leading direct marketing agency, and was Creative Director for the WiMAX Forum, a global Internet and telecom consortium. In the computer industry, Edward has developed corporate training solutions for Nike, Adidas, Columbia Sportswear, Kaiser Permanente, and FEI, among others. In addition, he has delivered professional services to NANOG, the Institute of Electrical and Electronics Engineers (IEEE), the Internet Corporation for Assigned Names and Numbers (ICANN), Facebook, Intel® and Mentor Graphics. Edward also developed the first web design program at the Pacific Northwest College of Art where he taught web and graphic design and interactive media courses for more than a dozen years. In his free time, Edward is involved in community theatre aimed at supporting local charities. His most recent production was playing the lead in the musical "Oklahoma!”
Valerie Wittkop: Valerie has worked with the North American Network Operators’ Group (NANOG) community since 2006 in various roles including support of the Program, Communication, and Development Committees, as well as Annual Elections. In 2015 she became the Program Director for NANOG and continues to work in that roll today. With a demonstrated history of working in the Internet industry, she is a business development professional with skills in non-profit management, fundraising, networking, sales, and start-ups. Valerie holds a BS focused in Business Management from University of Phoenix.
Speakers
Full Abstract

New technologies are emerging to improve the performance and scale of large cloud-centric data center networks as they evolve to address new and popular application scenarios such as Artificial Intelligence (AI) and Online Data Intensive services (OLDI). Some of these technologies are advancing through the standardization process and could soon be broadly available to industry. Other new technologies are in the research phase and would benefit from a review with operators. This track will overview the recent developments, particularly from the perspective of the IEEE 802 and IETF standards committees. It will also encourage the participants to discuss problems and inefficiencies in their current data center operations, trying to identify which problems are currently being addressed and which are in need of new solutions or further developments. The track is a followup to a talk at NANOG 75 ("Lossless Data Center Networks: Opportunities for NANOG Engagement with IEEE 802 Nendica") and will seek to lay the groundwork for future activity at NANOG 77, if deemed appropriate. The track will feature presentations by 3 speakers and Q&A with the attendees.

Speakers
Roger Marks, EthAirNet Associates
Roger Marks of EthAirNet Associates is Chair of the IEEE 802 “Network Enhancements for the Next Decade” Industry Connections Activity ("Nendica") within the IEEE 802.1 Working Group. He has served on the Executive Committee of the IEEE 802 LAN/MAN Standards Committee since 1998, currently as Second Vice Chair. He initiated the IEEE 802.16 Working Group and has served as its chair since inception. Marks received his A.B. degree from Princeton University and his Ph.D. from Yale University. He served as a physicist in the U.S. National Institute of Standards and Technology for 17 years. He is currently interested in novel solutions for data center networking.

Speakers
  • 3 Roger Marks - EthAirNet Associates
Full Abstract

RDMA is increasingly deployed in datacenter environments. Deployment introduces an additional level of fabric complexity in management and diagnosis. This talk will introduce operators to RDMA concepts from a Fabric management point of view, inclusive of caveats, quirks, misconceptions which have skewed deployment decisions. With multiple variants of RDMA at play (IB, iWARP, RoCE) and differing qualities/properties of each, from workload level capabilities to E2E fabric resiliency; a thorough understanding of each is needed for successful deployment.

Speakers
Omar Cardona, Microsoft
Omar Cardona is a Principal Software Engineering Lead and Architect at Microsoft. He drives the design and delivery of Windows virtual networking, accelerations, and offloads. Specifically, Server and Container network performance for Private and Public Cloud inclusive of scalability and diagnosis. Prior to Microsoft, Omar was a Software Architect at IBM focusing on Ethernet, RDMA, IO Virtualization, and Performance. Omar holds >60 patents and multiple publications in Core Systems and IO design.

Speakers
  • 3 Omar Cardona
Full Abstract

In this 90-minute session we will be tackling a major issue and it is NOT yet-another IPv6 exhaustion case. We understand making every single router dual stack is a real pain and can open the doors to new bugs, security vulnerabilities and entails a steep learning curve but still the IPv6 customers are knocking on our doors with urgent needs. So that, in this session we’ll demystify two little-known architectures namely 6PE and 6VPE and we will demonstrate how they can enable operators to keep their current infrastructure on IPv4 yet still support islands of IPv6 locally and globally in a quick and non-invasive fashion. This session is aimed at tearing down the gruesome walls between the traditional network operators and the world of IPv6 MPLS to satisfy the growing needs to IPv6 using what has been proven and known to the industry for over a decade. We’ll show that most operators are under negligible pressure to upgrade the cores and backbones in order to support IPv6. Yet, they are under tremendous pressure to support the new addressing on the edges of their networks. This could be a manufacturer needing IPv6 in their product testing labs, a software engineering company running IPv6 only for their development teams or even a small MPLS VPN service provider approached by customers about their IPv6 enabled sites or even access to the IPv6 Internet. PRESENTERS Kam Agahian (Qualcomm – Senior Manager of Network Engineering). CCIEx2 ; 20+ years of networking experience. Shahid Shafi (Qualcomm – Director of IT Networking). CCIEx3, 20+ years of networking experience.

Speakers
Kambiz Agahian

Full Abstract

Speakers
Paul Congdon, Tallac Networks
Paul Congdon is a co-founder and is the Chief Technology officer (CTO) of Tallac Networks. He has over 34 years of experience in the networking industry and has become a widely esteemed inventor and leader in the networking industry. Prior to Tallac Networks, Paul was a Fellow at Hewlett Packard’s Networking and Communications Labs with responsibility for HP’s research in mobility, wireless and SDN network infrastructure. Paul has led, chaired, and is currently contributing widely to industry standards in the IEEE and IETF. Paul has a PhD in Computer Science from the University of California, Davis.

Paul Congdon: Paul Congdon is a co-founder and is the Chief Technology Officer (CTO) for Tallac Networks. He has over 34 years of experience in the networking industry and has become a widely esteemed inventor and leader in the networking industry. Prior to Tallac Networks, Paul was a Fellow at Hewlett Packard Networking and Communications Labs with responsibility for HP’s research for mobility, wireless and SDN network infrastructure. Paul has led, chaired, and is currently contributing widely to industry standards in the IEEE and IETF. Paul is currently the IEEE 802.1 Maintenance Task Group Chair, responsible for the well being of all IEEE 802.1 published standards. Paul has a PhD in Computer Science from the University of California, Davis.
Speakers
  • 3 Paul Congdon - Tallac Networks
Full Abstract

In this 90-minute session we will be tackling a major issue and it is NOT yet-another IPv6 exhaustion case.
We understand making every single router dual stack is a real pain and can open the doors to new bugs, security vulnerabilities and entails a steep learning curve but still the IPv6 customers are knocking on our doors with urgent needs. So that, in this session we’ll demystify two little-known architectures namely 6PE and 6VPE and we will demonstrate how they can enable operators to keep their current infrastructure on IPv4 yet still support islands of IPv6 locally and globally in a quick and non-invasive fashion. This session is aimed at tearing down the gruesome walls between the traditional network operators and the world of IPv6 MPLS to satisfy the growing needs to IPv6 using what has been proven and known to the industry for over a decade. We’ll show that most operators are under negligible pressure to upgrade the cores and backbones in order to support IPv6. Yet, they are under tremendous pressure to support the new addressing on the edges of their networks. This could be a manufacturer needing IPv6 in their product testing labs, a software engineering company running IPv6 only for their development teams or even a small MPLS VPN service provider approached by customers about their IPv6 enabled sites or even access to the IPv6 Internet.

PRESENTERS
Kam Agahian (Qualcomm – Senior Manager of Network Engineering). CCIEx2 ; 20+ years of networking experience.
Shahid Shafi (Qualcomm – Director of IT Networking). CCIEx3, 20+ years of networking experience.

Kam Agahian: Kam is the director of cloud engineering with Oracle in Southern California with over 24 years of experience in designing and implementing complex network architectures. Over the years Kam has interviewed over 1000 candidates in North America, APAC and EMEA for various network engineering and leadership positions. Kam has previously presented at NANOG77 in Austin, TX and NANOG75 in Washington DC on IPv6 over MPLS and network engineering job interview processes. You can follow his random thoughts on Twitter.
Speakers
  • 3 Kam Agahian