^ Top

NANOG 68 Agenda

All times listed in Central Time.

**Agenda is subject to change.**

Webcast: http://www.verilan.com/clients/nanog/index.php

Floor Plans: Lobby Level, Banquet Level, Terrace Level

The NANOG 68 Agenda icalendar feed is now available! 
Download the .ics file and import into your calendar, or subscribe to the URL: 
http://www.nanog.org/sites/default/files/ical/nanog.ics  

 

Download the Guidebook app! If you already have the app, you can search for NANOG 68 starting October 6th and download the guide to your mobile device.

 

NANOG 68 Host Sponsor: CyrusOne

NANOG 68 Connectivity Sponsor: AT&T

Sunday, October 16 2016
Time/Webcast:Room:Topic/Abstract:Presenter/Sponsor:Presentation Files:
4:00pm - 6:00pmRegency Foyer (Banquet Level)Registration
Monday, October 17 2016
Time/Webcast:Room:Topic/Abstract:Presenter/Sponsor:Presentation Files:
8:30am - 4:30pmRegency Foyer (Banquet Level)Espresso BarSponsors:
8:30am - 11:00amGold Room & Parisian Room (Banquet Level)Extended BreakfastSponsors:
8:30am - 5:00pmRegency Foyer (Banquet Level)Registration
10:00am - 10:30am

Regency Ballroom (Banquet Level)

Conference Opening

Welcome to Chicago! A word from our local Host and Connectivity Sponsors as well as the NANOG Board and Program Committee Leadership.

View full abstract page.
Speakers:

  • L Sean Kennedy, XO Communications
  • L Sean Kennedy is an active member of the Internet Engineering community and a Director of Network Engineering at XO Communications. He has been a member of the NANOG Program Committee since 2013.

  • Josh Snowhorn, CyrusOne
  • Josh Snowhorn currently serves Vice President & GM of Interconnection at CyrusOne, and brings 16 years of data center and interconnection experience to his role, which is largely dedicated to expanding the CyrusOne National Internet Exchange (IX) across the company’s worldwide data center locations. Josh and his team have designed and implemented the CyrusOne National IX strategy to significantly increase the number of telecom and cloud providers available to customers, enabling Internet peering and leveraging industry-changing interconnection ecosystems. This proprietary and innovative technology platform was honored by DataCenter Dynamics as the 2014 Outsourced Innovation of the Year. Before joining CyrusOne in 2012, Josh spent the previous 12 years as Vice President at Terremark, a global Data Center and Cloud services provider acquired by Verizon in 2011. There he developed new business in emerging markets and drove all business related to peering and interconnection throughout the company’s portfolio. While at Terremark he co-founded the Global Peering Forum (GPF), a joint venture with Equinix, AMS-IX, DE-CIX, and LINX that continues to this day as an independent entity with many additional host partners. Josh serves on the Global Peering Forum Board of Directors, the Open-IX Board of Directors (term expires in 2016), the McLaren Car Club Board of Directors and the Austin Waldorf School Board of Trustees. He resides in Austin, Texas with his wife, two daughters, two dogs and a flock of chickens.

  • David Temkin, Netflix
  • Dave Temkin is the Director of Global Networks for Netflix. Having been hired to build the Open Connect CDN, he is responsible for all network architecture and strategy as well as the operations fo the Netflix network (AS2906). Before Netflix, he was at Yahoo!, where he focused on Layer 4-7 network architecture, having been brought in through the successful acquisition of Right Media where he was the Global Head of Networks. In his spare time he enjoys travel and philanthropy - both through volunteering at technical organizations such as NANOG, where he is vice chair, Open-IX, where he is the chairman and co-founder and FL-IX, where he is also chairman and cofounder, as well as on the board of Children of Bellevue.
youtubeConference Opening
pdfConference Opening(PDF)
10:30am - 11:30am

Regency Ballroom (Banquet Level)

Keynote: IANA Transition

Scott Bradner will discuss the history of Internet Governance leading up to the transition of oversight of the IANA function from NTIA to the internet's multistakeholder community.

View full abstract page.
Speakers:

  • Scott Bradner, Harvard University, retired
  • Scott Bradner was involved in the design, operation and use of data networks at Harvard University since the early days of the ARPANET. He was involved in the design of the original Harvard data networks, the Longwood Medical Area network (LMAnet) and New England Academic and Research Network (NEARnet). He was founding chair of the technical committees of LMAnet, NEARnet and the Corporation for Research and Enterprise Network (CoREN). Mr. Bradner retired from Harvard University in 2016 after 50 years working there in the areas of in computer programming, system management, networking, IT security and identity management. He still does some patent related consulting.
youtubeKeynote: IANA Transition
pdfKeynote: IANA Transition(PDF)
11:30am - 12:00pm

Regency Ballroom (Banquet Level)

Desperately Seeking Default

Is "default" the same all over the Internet? Does every component network in the Internet see the same of routes, or are there routes that are only visible to a subset of the Internet? This presentation analyses the route sets advertised to a number of the route collection points and looks for differences in the various route sets to see where and how "default" differs between networks.

View full abstract page.
Speakers:

  • Geoff Huston, APNIC
  • Geoff Huston works as the Chief Scientist at APNIC, the Regional Internet Registry for the Asia Pacific region. His research interests include addressing, routing, the domain name system, security and measurement.
youtubeDesperately Seeking Default
pdfDesperately Seeking Default(PDF)
12:00pm - 1:30pmGold Room (Banquet Level)Newcomers Lunch (Invitation Only)Sponsors:
12:00pm - 1:30pmInternational Ballroom (Lobby Level)Welcome LunchSponsors:
1:30pm - 2:00pm

Regency Ballroom (Banquet Level)

Routing protocol migrations in a live datacenter

As network topologies and protocols evolve over the years, it occasionally becomes apparent that a wholesale re-implementation of a datacenter's routing layer is required in order to continue to scale the network. And - often it’s necessary to “do it live!”. I present many lessons learned from Twitter’s migration from OSPF to BGP topologies in our datacenters - what went well, what didn’t go well, and what sort of challenges are encountered during a live datacenter migration.

View full abstract page.
Speakers:

  • Chris Woodfield, Twitter
  • Chris Woodfield is a Senior Staff Network Engineer at Twitter, where he concentrates on datacenter and edge network and traffic engineering. Prior to Twitter, he had roles at Yahoo! and Internap. He is a member of the NANOG Program Committee and is running for a seat on the ARIN Advisory Council in the upcoming election.
youtubeRouting protocol migrations in a live datacenter
pdfRouting protocol migrations in a live datacenter(PDF)
2:00pm - 3:00pm

Regency Ballroom (Banquet Level)

Board Candidate Presentations

A time to meet the candidates on the 2016 Board of Directors Ballot, hear their presentations, and ask them questions.

View full abstract page.
Speakers:
  • Ryan Donnelly, NANOG Board of Directors.
youtubeBoard Candidate Presentations
3:00pm - 3:30pmRegency Foyer (Banquet Level)BreakSponsors:
3:30pm - 4:00pm

Regency Ballroom (Banquet Level)

Network Automation with Salt and NAPALM

It is already well known that Salt is a very powerful and flexible system. At CloudFlare it also proved to be extremely scalable, being able to manage hundreds of servers. The new feature of Proxy Minions come with many advantages and if used with a proper library to interact with the network device, provides almost unlimited facilities for the network teams to control their equipment. Beginning with version 2016.3, NAPALM will be integrated in Salt, which means even more flexibility and more ease in configuration than anything else.

View full abstract page.
Speakers:

  • Mircea Ulinic, Cloudflare
  • Mircea is one of Cloudflare's network engineers, spending most of his time writing code for network automation. Sometimes he likes to talk about the stuff he's working on and how automation really helps to maintain reliable, stable and self-resilient one of the biggest global networks. Previously, he was involved in research and later worked for EPFL in Switzerland and an European service provider based in France. Besides networking, he has a strong passion for radio communications (especially mobile networks), mathematics and physics.
youtubeNetwork Automation with Salt and NAPALM
pdfNetwork Automation with Salt and NAPALM(PDF)
4:00pm - 4:40pm

Regency Ballroom (Banquet Level)

Internet-scale virtual networking with ILA

In this talk we describe Facebook's ongoing experience deploying the Identifier-Locator Addressing (ILA) framework: an IPv6-only model for implementing per-process, location-independent addressing based on concepts originally developed in ILNP. The major benefits include logical addressing of application containers decoupled from physical infrastructure and the ability to uniquely name any process in the network. We present the motivation and concepts behind the ILA data and control planes and summarize our experience deploying it in the production, describing how we leverage the recent development in Linux kernel such as eBPF and XDP.

View full abstract page.
Speakers:

  • Petr Lapukhov, Facebook
  • Petr Lapukhov is a Network Engineer at Facebook, primarily focused on building and supporting software systems for network monitoring, optimization and control. Prior to Facebook he worked at Microsoft on various aspects of data-center design and deployment.
youtubeInternet-scale virtual networking with ILA
pdfInternet-scale virtual networking with ILA(PDF)
4:40pm - 5:00pm

Regency Ballroom (Banquet Level)

Partially FIBing

High-capacity attachment at large Internet exchange points would traditionally be done on a device capable of absorbing the fully available selection of routes. Not only is that not necessary, but for some kinds of applications (CDNs), nameservers, many classes of server, or extensions of existing networks, very minimal FIB routes are required on a switch or router. This provides an opportunity for reduced attachment cost, as well as minimal threat of obsolescence due to table size.

View full abstract page.
Speakers:

  • Joel Jaeggli
  • Joel Jaeggli is a principle network engineer at Fastly. He also serves as one of the directors of the IETF Operations and Management Area.
youtubePartially FIBing
pdfPartially FIBing (PDF)
5:00pm - 5:30pm

Regency Ballroom (Banquet Level)

Dispatches from the DNS Frontier

DNS-OARC regularly co-locates its Domain Name Systems Operations Workshops with NANOG, and is doing so again in Dallas immediately before NANOG68. The impact of DNS operational practice, infrastructure and abuse is fundamental to Internet Operations, yet often arcane and poorly understood. OARC seeks to bridge this gap through knowledge sharing, data gathering/analysis, community building, and outreach. This talk will give an introduction and summary for the wider NANOG audience of the latest DNS material presented at OARC25. The exact details will to be determined as the OARC25 Agenda firms up (see https://indico.dns-oarc.net/event/25/call-for-abstracts/ - planned publication date 1st September), but will include a pick of the best new DNS technology deployment experiences, best practices, data, and analysis, with pointers to sources of further information. OARC25 abstracts submitted to date include DNSSEC and TLD transfer experiences, Root key rollover, data format migration, and attack analysis, and there will be particular focus on resolver operations.

View full abstract page.
Speakers:

  • Paul Ebersman, Comcast
  • 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 Comcast 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, working for various internet infrastructure companies including ISC and Nominum before coming to Comcast.
youtubeDispatches from the DNS Frontier
pdfDispatches from the DNS Frontier(PDF)
5:30pm - 7:00pmInternational Ballroom (Lobby Level)Peering PersonalsSponsors:
7:00pm - 10:00pmOffsite: Perot MuseumSocial
Additional information
Sponsors:
Tuesday, October 18 2016
Time/Webcast:Room:Topic/Abstract:Presenter/Sponsor:Presentation Files:
8:30am - 4:30pmRegency Foyer (Banquet Level)Espresso BarSponsors:
8:30am - 9:30amGold Room (Banquet Level)Members Breakfast (Invitation Only)
8:30am - 9:30amInternational Ballroom (Lobby Level)Power BreakfastSponsors:
8:30am - 5:00pmRegency Foyer (Banquet Level)Registration
9:30am - 10:15am

Regency Ballroom (Banquet Level)

100G+ Data-Center Evolution and Challenges

Data centers around the globe are migrating to fiber. The move to cloud services, massive virtualization and explosive growth in data have transformed the business role of data centers. This presentation will cover the following topics in detail: • The trends and evolution of the data-center market. • The main physical-layer infrastructures that can be leveraged to build next-generation data centers. • Why connector insertion and return losses are increasingly important for migration to 100G+ interfaces. • The most relevant test and monitoring steps from construction to turn-up and troubleshooting, for 100G+ interfaces such as CFP4 and QSFP28, and network performance tests in the intra- and interconnect. • Service assurance possibilities offered to the DC industry, including infrastructure monitoring, cloud monitoring and subscriber analytics.

View full abstract page.
Speakers:

  • Nicholas Gagnon, EXFO
  • At EXFO, Nicholas is responsible for business development in Data center market segment, worldwide. He is working with each regional teams to develop new test solutions and communicate the value of the existing one. Prior to joining EXFO, he held different sales and marketing roles in Telecom / Datacom test solutions, Biophotonics industries.
youtube100G+ Data-Center Evolution and Challenges
pdf100G+ Data-Center Evolution and Challenges(PDF)
9:30am - 10:15am

Venetian (Lobby Level)

BGP- The High Way of Internet

This tech talk is based on the recent & future advancements in BGP. BGP is so successful in gluing together the Internet because of its simplicity. Multimedia, Mobile Internet and Cloud Services has generated massive bandwidth explosion in recent times & is expected to grow exponentially in future. Service providers are looking at rapid service deployment based on BGP to cater to various customers. Evolution of BGP based services need to go from basic technologies to very advanced infrastructures to meet the future goals. This tech talk focuses on such BGP advancements which would help the providers to rapidly deploy new services, better manage their infrastructure, provide network visibility and other benefits such as traffic engineering, SDN based WAN deployments etc. The discussion would be around the below area: BGP infrastructure advancement VPN enhancement High Availability Route Reflection advancement BGP Multipath BGP enabler for SDN

View full abstract page.
Speakers:

  • Rohit Bothra, Brocade Communications
  • Rohit Bothra is a Staff Engineer at Vyatta - a Brocade company, specializing in virtual network appliances, system & solution testing, performance & scale measurement of virtual systems. He has over nine years of rich experience in Networking industry. He has worked with many major service providers in the Asia Pacific region in the area of IP, IPv6, MPLS. His area of expertise includes NFV, VNFs, Routing Protocols, MPLS, IPv6, HA Systems, Routing Platforms, Network Security, Network Operations, providing Network deployment solutions to different service provider customers. In the past, he has represented Cisco at APNIC conferences. He is passionate about learning new technologies.
youtubeBGP- The High Way of Internet
pdfBGP- The High Way of Internet(PDF)
10:15am - 11:00am

Regency Ballroom (Banquet Level)

PERISCOPE: Standardizing and Orchestrating Looking Glass Querying

Looking glasses (LG) servers enhance our visibility into Internet connectivity and performance by offering a set of distributed vantage points that allow both data plane and control plane measurements. However, the lack of input and output standardization and limitations in querying frequency have hindered the development of automated measurement tools that would allow systematic use of LGs. In this paper we introduce Periscope, a publicly-accessible overlay that unifies LGs into a single platform and automates the discovery and use of LG capabilities. Periscope can handle large bursts of requests, with an intelligent controller coordinating multiple concurrent user queries without violating the various LG querying rate limitations. I show that Periscope significantly extends our view of Internet topology with over 2,000 vantage points in 478 cities.

View full abstract page.
Speakers:

  • Vasileios Giotsas, UCSD/CAIDA
  • I'm a researcher at CAIDA. My interests include the measurement and analysis of the Internet interconnection and peering ecosystem at different levels of granularity, the resilience of the critical Internet infrastructures during outages and attacks, and the development of tools that facilitate the research and engineering communities. I earned my Ph.D in Communications Networks and my Master in Data Communications, Networks and Distributed Systems from University College London (UCL), and my Bachelor in Computer Science from the University of Piraeus in Greece.
youtubePERISCOPE: Standardizing and Orchestrating Looking Glass Querying
pdfPERISCOPE: Standardizing and Orchestrating Looking Glass Querying(PDF)
10:15am - 11:00am

Venetian (Lobby Level)

Upcoming Changes to BGP Flow Specification

Flow Specification insert policies that control and channel traffic flows in networks to orchestrate traffic in SDN, direct traffic to pathways for specific processes in NFV networks, and prevent distributed denial of service attacks (DDoS) within and between networks. Flow policy can be set in devices via: 1) statically configured in policy routing, 2) reboot ephemeral configuration, and 3) protocol session ephemeral state (BGP-FS and ISIS-FS). This talk provides an overview of flow policy, describes the upcoming change to BGP Flow Specification policy being specified by IETF, and ask NANOG’s input on these changes. In order to help NANOG provide input, this talk accompanied by open source demo code that sets flow policy via static configuration, I2RS protocol, and BGP protocol session state.

View full abstract page.
Speakers:

  • Susan Hares, Huawei
  • Susan Hares ([email protected]) has over 30 years of experience in International Standard Organizations for Internet and IT technology (IETF, IEEE, BBF, MEF, MAP/TOP) that use consensus decision making to create standards. From 1995-2007, Susan founded and was the CTO for NextHop Technologies, a company developing network hypervisor and routing/switching software suites. From 2008-2009, Susan managed a team at Green Hills Software developing routing software in a secure hypervisor that launched early versions of cloud software. From 2010-2013, Susan Hares was a Senior Director of the IPBU Standards Team in the Future US R&D division of Huawei, a Chinese telecommunications company. Since 2014, Susan has been a Consultant working directly with Huawei’s Beijing Research Center. Susan currently chairs the following IETF working groups: IDR (bgp), I2RS, and TRILL. Susan is Ph.D. student at Regent University Business and Leadership School in Global Leadership.
youtubeUpcoming Changes to BGP Flow Specification
pdfUpcoming Changes to BGP Flow Specification(PDF)
11:00am - 11:30amRegency Foyer (Banquet Level)BreakSponsors:
11:30am - 12:15pm

Venetian (Lobby Level)

25-50-100G Ethernet Options and Experience in the Datacenter

There are rapid changes occurring in the intra-data center connectivity realm. We took interest in work from the 25-50-100 Gigabit Ethernet camp. The connectivity options are now practical choices for networking inside the datacenter. In this talk, I will review options for 25-50-100G interfaces between datacenter switches. I will use a recent project at LinkedIn to show selection process, planning concepts and lessons learned with one of those available 100G modules.

View full abstract page.
Speakers:

  • Paul Zugnoni, LinkedIn
  • Paul Zugnoni is a network architect at LinkedIn, supporting datacenter and backbone networks. In his 15+ years of professional experience he has led datacenter-based network solutions for online retail and trading customers at Verizon Business, managed network design and operations for a real-time, high-bandwidth, low-latency video delivery network at OnLive, and led Edge POP and APAC datacenter build projects at LinkedIn. https://www.linkedin.com/in/paulzugnoni/
youtube25-50-100G Ethernet Options and Experience in the Datacenter
pdf25-50-100G Ethernet Options and Experience in the Datacenter(PDF)
11:30am - 1:00pm

Regency Ballroom (Banquet Level)

Security Track

The NANOG security track explores the latest in current network security threats, defenses and research. Contact the track coordinator to reserve some time to lead a discussion or present a topic of interest. * DDoSMON, presented by 360.cn * DNS Exfiltration, presented by Rackspace * Spoofer Project, presented by CAIDA

View full abstract page.
Moderators:

  • Jesse Sowell, Standford University
  • Jesse Sowell is a Research Affiliate at MIT and runs DevOps and Measurement Visualization at the Markley Group. Jesse's PhD research was a study of institutional mechanisms and attendant forms of authority that support Internet infrastructure operations. In other words, what are the characteristics of the social glue that ensures interconnection among a private network of networks continues to cohere into the modern Internet? Jesse's current research interests comprise cost and risk models of Internet infrastructure, incentives in malware value networks, and the political challenges of developing joint capabilities between operational communities and conventional governance actors.
Panelists:
  • Nolan Berry.
  • k claffy, CAIDA
  • Kimberly Claffy ("kc claffy") is founder and director of the Center for Applied Internet Data Analysis (CAIDA), a resident research scientist of the San Diego Supercomputer Center at UC, San Diego, and an Adjunct Professor in the Computer Science and Engineering Department at UC, San Diego. Her research interests span Internet topology, routing, security, economics, future Internet architectures, and policy. She leads CAIDA research and infrastructure efforts in Internet cartography, aimed at characterizing the changing nature of the Internet's topology, routing and traffic dynamics, and investigating the implications of these changes on network science, architecture, infrastructure security and stability, and public policy. She has been at SDSC since 1991 and holds a Ph.D. in Computer Science from UC San Diego.
  • Cory Schwartz.
Speakers:
  • Yiming Gong.
  • Qiang Ke, Qihoo 360.
  • John Kristoff, DePaul University
  • John Kristoff is currently employed by DePaul University as a network architect and adjunct faculty. In the past, he has held positions with Team Cymru, the Dragon Research Group, UltraDNS/Neustar and Northwestern University. John is active in a variety of communities including NANOG, IETF, FIRST, ops-trust, nsp-security, Internet2 and REN-ISAC.
youtubeSecurity Track
pdfSecurity Track(PDF)
pdfSecurity Track(PDF)
pdfSecurity Track(PDF)
12:15pm - 1:00pm

Venetian (Lobby Level)

Case Study: T-Mobile Network Design Evolution

In this session, we will describe the main design model and components used to build the T-Mobile USA network, present major evolutions in the last ten years such as architecture, technology, and routing protocol changes. Finally, some lessons and recommendations learned from our experience will be shared. Hierarchical Network Design Model T-Mobile Network at a glance T-Mobile Design Model Evolution Architecture Evolution Technology Evolution Routing Protocol Evolution BGP RR Evolution Lessons Learned and Recommendation

View full abstract page.
Speakers:

  • ISSA ABU EID
  • Issa AbuEid is a principal engineer with Enterprise IP and Security Strategy team at T-Mobile USA. He joined T-Mobile in June 2007 as a senior engineer and prompted to a principal engineer in 2010. In addition to a fifteen years of network architecture and engineering experience toward designing, implementing, and support networks, he holds a master’s degree in Computer Information System as well as three CCIE in Routing and Switching, Service Provider, and Data Centers.
pdfCase Study: T-Mobile Network Design Evolution(PDF)
youtubeCase Study: T-Mobile Network Design Evolution
1:00pm - 2:30pmInternational Ballroom (Lobby Level)LunchSponsors:
2:30pm - 3:00pm

Regency Ballroom (Banquet Level)

First Steps in Bufferbloat Mitigation

The Cable Industry has been investigating techniques for mitigating the Bufferbloat (www.bufferbloat.net) condition seen on the Internet. CableLabs has selected an Active Queue Management (AQM) technique called PIE (Proportional Integral controller Enhanced) as part of the DOCSIS 3.1 standards. In advance of AQM capable DOCSIS 3.1 equipment becoming available, Comcast has conducted a field trial using a static queue management approach. The goal of the trial was to measure the impact of a bufferbloat mitigation technique implemented on Comcast leased DOCSIS 3.0 modems. This presentation shares the trial methodology, results obtained, and success & challenges encountered.

View full abstract page.
Speakers:

  • Carl Klatsky, Comcast
  • Carl Klatsky currently works for Comcast as a Senior Principal Engineer. He has led critical technology changes to Comcast's network, such as transitioning Comcast's Voice CPE from using the NCS protocol to the SIP protocol, developing & deploying DOCSIS 3.0 CPE, transitioning Comcast's Voice CPE from IPv4 to IPv6, and his most recent work in evaluating bufferbloat mitigigation techniques on DOCSIS CPE. Carl received his M.S. in Software Engineering from Monmouth University and his B.E. in Electrical Engineering from Stevens Institute of Technology.
youtubeFirst Steps in Bufferbloat Mitigation
pdfFirst Steps in Bufferbloat Mitigation(PDF)
3:00pm - 3:45pm

Regency Ballroom (Banquet Level)

The Current Economics of Cyber Attacks

Often we discuss the changing threat landscape from a pure technical or vulnerability perspective, however this does an injustice to element of ease, cost and access to attacks. This presentation will provide attendees with an up-to-date picture of the rapidly changing landscape of attack tools and services. In addition, the presentation will provide an understanding of how the combination of the proliferation of these tools and their corresponding use has dramatically changed the dynamics of the return on defense strategies. This presentation will provide unique insight into the world of the Darknet, specific customer attack stories, new economic models of measuring security deployments and a refreshed look at how controls should be deployed going forward.

View full abstract page.
Speakers:

  • Ron Winward, Radware
  • Ron Winward is a Security Evangelist for Radware, where he helps execute the company’s thought leadership on today’s security threat landscape. Ron brings nearly 20 years of experience in the Internet service provider space, most recently as Director of Network Engineering for a global infrastructure and colocation provider. With an expertise in network architectures and DDoS mitigation, Ron has helped design solutions for carriers, enterprises, and cybersecurity service providers around the world.
youtubeThe Current Economics of Cyber Attacks
pdfThe Current Economics of Cyber Attacks(PDF)
3:45pm - 4:15pm

Regency Ballroom (Banquet Level)

Decoding Performance Data from Large-Scale Internet Outages

Large-scale Internet events create complex, and often confounding data. This talk will explore three recent events--the June 2016 DNS root server DDoS, the May 2016 Sea-Me-We-4 cable cut and the April 2016 AWS route leak--that had large effects on Internet infrastructure and availability. We’ll discuss performance data associated with these events and approaches to detecting and troubleshooting them.

View full abstract page.
Speakers:

  • Mohit Lad, ThousandEyes
  • Mohit shapes and executes ThousandEyes' vision. Mohit brings over 10 years of experience in designing and implementing systems to solve hard networking problems. Mohit received a Ph.D. in Computer Science from UCLA where he conducted research in the area of large scale network diagnostics. Prior to ThousandEyes, Mohit worked at Nokia where he built a system to monitor mobile services performance. Mohit is an active member of North American Network Operations Group (NANOG) and has served on its Program Committee for 4 years from 2009 to 2013.
youtubeDecoding Performance Data from Large-Scale Internet Outages
pdfDecoding Performance Data from Large-Scale Internet Outages(PDF)
4:15pm - 4:45pmRegency Foyer (Banquet Level)BreakSponsors:
4:45pm - 5:30pm

Regency Ballroom (Banquet Level)Lightning TalkyoutubeLightning Talk
pdfLightning Talk(PDF)
pdfLightning Talk(PDF)
pdfLightning Talk(PDF)
pdfLightning Talk(PDF)
5:30pm - 6:00pm

Regency Ballroom (Banquet Level)

BackConnect’s Suspicious BGP Hijacks

In early September 2016, security blogger Brian Krebs broke a story about an Israeli DDoS-for-hire service, vDOS, which had been hacked, revealing “tens of thousands of paying customers and their (DDoS) targets.” Afterwards, Krebs noticed that vDOS itself was also a victim of a recent BGP hijack from a company called BackConnect. The CEO of BackConnect defended this act as justifiable and said it was a one-time event. Krebs then contacted Dyn for some assistance in researching what appeared to be a series of BGP hijacks conducted by BackConnect over the past year. What emerges from this analysis is that the hijack against vDOS probably wasn’t the first one conducted by BackConnect. This talk will review multiple incidents where it appears that BackConnect used BGP hijacks and, via the use of forged AS paths, sometimes obscured their involvement in this activity. Separately, this raises the philosophical question of whether there could be justification for a "defensive" BGP hijack. This talk will draw on the analysis in the following blog posts: http://research.dyn.com/2016/09/backconnects-suspicious-bgp-hijacks/ http://krebsonsecurity.com/2016/09/ddos-mitigation-firm-has-history-of-hijacks/

View full abstract page.
Speakers:

  • Doug Madory, Dyn
  • Doug Madory is the Director of Internet Analysis at Dyn (formerly Renesys) where he works on Internet infrastructure analysis. In a recent profile, The Washington Post dubbed him “The Man who can see the Internet" for his reputation in identifying significant developments in the global layout of the Internet. Doug is regularly quoted by major news outlets (including The New York Times, NPR's All Things Considered, and NBC Evening News) about developments ranging from national Internet blackouts to BGP hijacks to transoceanic submarine cables. Prior to Dyn, Doug held positions such as chief of computer security for Dartmouth-Hitchcock Medical Center, senior research engineer at BAE Systems, and communications officer in the US Air Force. He holds computer engineering degrees from the University of Virginia and Dartmouth College.
youtubeBackConnect’s Suspicious BGP Hijacks
pdfBackConnect’s Suspicious BGP Hijacks(PDF)
6:00pm - 8:00pmInternational Ballroom (Lobby Level)Beer n GearSponsors:
9:00pm - 11:55pmOffsite: The Happiest HourSocial
Additional information
Sponsors:
Wednesday, October 19 2016
Time/Webcast:Room:Topic/Abstract:Presenter/Sponsor:Presentation Files:
8:30am - 4:30pmRegency Foyer (Banquet Level)Espresso BarSponsors:
8:30am - 9:30amInternational Ballroom (Lobby Level)Power Breakfast
8:30am - 5:00pmRegency Foyer (Banquet Level)Registration
9:30am - 10:00am

Regency Ballroom (Banquet Level)

Being Open: How Facebook Got Its Edge

Learn about how Facebook's engineers have managed enormous waves of growth while evolving their edge networks to connect more than 1.6 billion people. Scaling is a challenge for any network; but Facebook's rapid growth, both virtually online and physically in their peering networks, is something few have ever seen. In this session, you'll learn about how Facebook's engineers have evolved and grown their interconnections across the world, building vast volumes of peering capacity, as well as creating the automation to squeeze every last bit of efficiency and performance for the benefit of all the people who communicate through Facebook.

View full abstract page.
Speakers:

  • James Quinn, Facebook
  • James Quinn is a Network Engineer at Facebook. He focuses primarily on the growth and development of Facebook's edge and backbone networks. Before Facebook, he worked for many years at Juniper Networks in several roles focused on the growth of mobile networks and network security.
youtubeBeing Open: How Facebook Got Its Edge
pdfBeing Open: How Facebook Got Its Edge(PDF)
9:30am - 11:00am

Venetian (Lobby Level)

Ok, We Got YANG Data Models, Now What?

This session describes how to significantly simplify network programmability using APIs generated from YANG data models. Model-driven APIs allow the network programmer to focus on the underlying structure of the device configuration and operational data. They abstract protocols, transports and encodings, plus they free the programmer from having to master the specifics of YANG. This session will show you how to get started with Python model-driven APIs using two open source projects: YDK-Py and YDK-gen. The session will end with a demonstration of the simple Python script using model-driven APIs to configure a BGP router using the OpenConfig BGP data model.

View full abstract page.
Speakers:

  • Santiago Alvarez
  • Santiago is a distinguished engineer at Cisco Systems focused on network routing and programmability. He is responsible for influencing technology innovation and driving its adoption worldwide. Santiago is a regular speaker at various networking conferences throughout the world and at Cisco Live. He is the author of the Cisco Press “QoS for IP/MPLS Networks”. Santiago holds a BS in Computer Science, a MS in Computer Science and a MS in Telecommunications.
pdfOk, We Got YANG Data Models, Now What?(PDF)
youtubeOk, We Got YANG Data Models, Now What?
10:00am - 10:30am

Regency Ballroom (Banquet Level)

Large BGP Communities

Large BGP Communities are a novel way to signal information between networks. Large BGP Communities are easy to use, implement and deploy. An example of a Large BGP Communities is: 2914:65400:38016. Large BGP Communities are composed of three 32-bit integers, separated by a colon. This is easy to remember and accomodates advanced routing policies that support 32-bit ASNs.

View full abstract page.
Speakers:

  • Greg Hankins, Nokia
  • Greg Hankins has been attending NANOG since 1998, first as a network operator and now as a hardware vendor. He also attends APRICOT, Euro-IX, various Peering Forums, RIPE, and regional operator conferences where he frequently speaks on network technology and operational topics. Greg currently works as a Senior Product Manager for Nokia.

  • Job Snijders, NTT
  • Job is actively involved in the Internet community both in an operational capacity and as a founder of cooperation efforts such as the NLNOG RING. He has taught service providers in the Middle East how to deploy IPv6 and has a passion for Routing Security and Automation. Job holds a position at NTT Communications' IP Development Department.
pdfLarge BGP Communities(PDF)
youtubeLarge BGP Communities
10:30am - 11:00am

Regency Ballroom (Banquet Level)Lightning TalkpdfLightning Talk(PDF)
pdfLightning Talk(PDF)
pdfLightning Talk(PDF)
youtubeLightning Talk
11:00am - 11:30amRegency Foyer (Banquet Level)BreakSponsors:
11:30am - 12:00pm

Regency Ballroom (Banquet Level)

VxLAN BGP-EVPN

VxLAN is one of the more important technologies for the modern day data centers but faces some challenges with its flood and learn mechanism. This session focuses on understanding the control plane architecture for VxLAN BGP-EVPN solution that helps overcome some of these challenges and help scale the VxLAN deployment. This session focuses on understanding the underlay, the overlay and the use of BGP-EVPN feature to advertise the MAC and MAC+IP information across the cloud.

View full abstract page.
Speakers:
  • Vinit Jain, Cisco Systems Inc.
pdfVxLAN BGP-EVPN(PDF)
youtubeVxLAN BGP-EVPN
12:00pm - 12:30pm

Regency Ballroom (Banquet Level)

21st Century iBGP Route Reflection

This presentation looks at how a large scale network operator in Africa has deployed carrier-grade route reflectors on commodity server hardware to support high end routing in its core network, made up of multiple PoP's spread across different countries in Africa and Europe.

View full abstract page.
Speakers:

  • Mark Tinka, SEACOM
  • Mark has been in the Internet industry since 1999, having helped build various Internet Exchange Points (IXP's) in Uganda, Swaziland, Zimbabwe, Zambia, Botswana and Malawi. He was also instrumental in shaping the technological advancements and construction of the service networks of Africa Online, a Pan-African Internet Service Provider with eight operations on the continent, Global Transit and TIME dotCom, two Malaysian sister organizations engaged in wholesale and retail infrastructure services in Malaysia, when he worked there. Mark is heavily involved in various community organizations such as AfNOG, AIS, APRICOT, MyNOG, AfPIF, SAFNOG, e.t.c. He is now based in Johannesburg, South Africa, where he is the Head of Engineering at SEACOM, a submarine cable operator along the coast of East and Southern Africa.
pdf21st Century iBGP Route Reflection(PDF)
youtube21st Century iBGP Route Reflection
12:30pm - 12:55pm

Regency Ballroom (Banquet Level)

Network Automation with State Machines

Automation has become vital to building large scale networks. However, building these networks and managing their entire life cycle with minimal human intervention remains a challenge. We realize that the fundamental action in this automation can be abstracted as reconciling the difference between the actual state and the desired state of the system. Guided by state machines, we have implemented a fully automated system to provision, turn-up, and manage our data center networks at Yahoo. In this system, the network architecture is modeled as a set of configuration templates. A no touch configuration generating engine is built on top of the model. The actual state of devices within the system is collected in real time by agents running on the devices. Additional data is pulled in from external sources, such as inventory databases, to feed the templating engine. Changes to the desired state come from input by engineers via the API, as well as state data collected from the devices. These changes then trigger the model to generate desired configurations for devices. Once a new version of a configuration has been generated, it advances through 3 states, GENERATED, RELEASED, and VALIDATED. These states are used to track the progress of a change and control the rate and sequence at which new configurations are released out into the network. The transition from GENERATED to RELEASED is where the rate and sequence of such changes are controlled, and will be explored in depth as part of this presentation. Once a configuration is in the RELEASED state, it is ready to be picked up by the network device. The device will then apply the configuration, run a series of health checks, and report the version of the active configuration to the system. This presentation will cover the overall design of the system, share the details of the state machines, walk through a specific use case, and discuss challenges faced when implementing the system.

View full abstract page.
Speakers:

  • Zoe Blevins
  • Zoe Blevins is an automation guru in the network team at Yahoo. Over the last 7 years, she has designed, implemented, and operated software systems to provision, manage, monitor and audit Yahoo’s global network. She is a CCNP holder, proficient in numerous programming languages, and a multi-time hackovation winner. Her primary focus over the last year has been designing and implementing a network management system that removes the need for operators to interact directly with network gear for day to day operations, and she is eager to share her findings with the rest of the industry.

  • 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.
pdfNetwork Automation with State Machines(PDF)
youtubeNetwork Automation with State Machines
12:55pm - 1:00pm

Regency Ballroom (Banquet Level)Announcement of Election/Bylaw ResultsSpeakers:
  • Ryan Donnelly, NANOG Board of Directors.
1:00pm - 2:30pm Lunch (On Your Own)
2:30pm - 3:00pm

Regency Ballroom (Banquet Level)

How to Navigate Getting IPv4 Space in a Post-Run-Out World

RIRs have traditionally been the primary source of both IPv4 and IPv6 addresses. However, in September of 2015 ARIN’s free pool of IPv4 addresses was depleted. RIPE and APNIC have both reached IPv4 exhaustion as well. Yet many organizations, even many of those that have already implemented IPv6, continue to need additional IPv4 addresses to run their business. While an IPv4 market has developed to fill this need that market can be difficult to navigate. It lacks transparency. Pricing fluctuates. RIR regulations surrounding transfers change regularly and are often not well understood by operators. At the same time stories have surfaced of registration hijacking and other issues that increase the level of risk involved in participating in the v4 market. This talk aims to provide guidance to organizations looking to navigate the process of acquiring IPv4 addresses by providing a set of best practices and guidance on things like vetting potential sellers, RIR transfer policies to be aware of when planning a purchase and transfer, structuring contracts to minimize risks, preparing successful RIR transfer requests and smoothly navigating the RIR transfer process.

View full abstract page.
Speakers:

  • Amy Potter, Hilco Streambank
  • Amy Potter is the Vice President of Hilco Streambank, where she heads up Hilco Streambank's IPv4 address brokerage business. Amy works with ISPs, hosting companies, datacenters and others to acquire IPv4 address space and navigate RIR transfer policies. Amy led the creation of ipv4auctions.com in order to increase transparency and efficiency in the IPv4 market. Amy is interested in lending her expertise in the IPv4 market to help create policies that make sense in a post-IPv4 exhaustion world. Amy has a law degree from the University of Notre Dame, and a Bachelor's degree from Santa Clara University. She was elected to the AC in October of 2015. Her current term expires 31 December 2018.
pdfHow to Navigate Getting IPv4 Space in a Post-Run-Out World(PDF)
youtubeHow to Navigate Getting IPv4 Space in a Post-Run-Out World
3:00pm - 3:30pm

Regency Ballroom (Banquet Level)

Smarter Purchasing of IPv4 Addresses in the Market

It may be easy to go out and find IPv4 addresses to buy. But there are a lot of nuances to buying space well. In this presentation, I will describe the steps I think all purchasers should take (regardless of continent/legal environment of the countries involved), and share some nuances that not everyone is aware of. Special emphasis on buying addresses for use in China, and detecting and avoiding fraud in both the market and at the RIRs.

View full abstract page.
Speakers:

  • David Huberman
  • David Huberman runs the IP addressing shop for Oracle's Bare Metal Cloud product. Prior to joining Oracle, David worked at Microsoft, Global Crossing, Telocity, and PBS. David also spent 10 years as a staffer at ARIN, and is a current member of the ARIN Advisory Council. David has a Bachelor of Arts degree in both Geography and Telecommunications from Indiana University.
pdfSmarter Purchasing of IPv4 Addresses in the Market(PDF)
youtubeSmarter Purchasing of IPv4 Addresses in the Market
3:30pm - 4:15pm

Regency Ballroom (Banquet Level)

Keynote: Internet of Things

Users will expect reliability, safety, privacy, ease of use, scalability, and autonomy (ie, they work even when NOT connected to the public Internet). There are many challenges ahead to deal with interoperability among the same of different brands of IOT. Most users would be unhappy with one app per device. When guests come and go, how hard or easy is it to grant them access to some or all of the household devices? And to disable their access when they leave? Under some conditions, policy, fire and EMT teams might need emergency access but only during the emergency. I won't have a lot of answers in this talk, just a lot of questions.

View full abstract page.
Speakers:

  • Vinton G. Cerf, Google
pdfKeynote: Internet of Things(PDF)
youtubeKeynote: Internet of Things
4:15pm - 4:45pmRegency Foyer (Banquet Level)Break
4:45pm - 5:15pm

Regency Ballroom (Banquet Level)

Rolling the Root Zone DNSSEC Key Signing Key

IANA/ICANN/PTI has prepared a plan to change the Root Zone DNSSEC KSK. This talk will present highlights of the plan to explain what will be happening, why this is important to follow, and what the audience needs to do as a result.

View full abstract page.
Speakers:

  • Edward Lewis, ICANN
  • Edward is a Senior Technologist in the Office of the CTO. Prior to joining ICANN he worked 11 years inside Internet registries of many types - gTLD, ccTLD, sTLD and RIR. He worked for a DNS hosting company. He co-chaired the original IETF WG that developed EPP. He developed some of the first DNSSEC codebases under the original DARPA contract in the 1990's. Before that, he worked building research networks attached to the NASA Science Internet (one of the three original backbones) and taught Networking courses at the University of Maryland - Baltimore County.
pdfRolling the Root Zone DNSSEC Key Signing Key(PDF)
youtubeRolling the Root Zone DNSSEC Key Signing Key
5:15pm - 5:45pm

Regency Ballroom (Banquet Level)

Test your way to a better deployment!

Ever wondered why your network deployment timelines always get pushed out? Or why your network config validation takes weeks, sometimes months? Could you enable better automation, increased access to test-networks for testers, detailed comprehensive reports without hiring an army of developers? In this session we introduce a whole new way to test network devices that leverages test cases from the open source community and enables automated network testing at scale on virtual hardware. We introduce a new open-source library for the RobotFramework that enables coders and network engineers to represent and share complicated test cases around BGP, OSPF, Streaming Telemetry and more, in simple text files. All of this is achieved with the help of vendor agnostic APIs using tools like Napalm, ydk-py and others in a pluggable format within the library. We will also showcase how such a framework enables devops workflows to percolate through the organization stack and increases code quality across tools and core network operating software.

View full abstract page.
Speakers:

  • Akshat Sharma, Cisco Systems Inc.
  • Akshat Sharma is a Technical Marketing Engineer in the Web Solutions space at Cisco .His day job includes a core focus on Network operating system architecture and Devops tooling in the Web and Service Provider domain. With 7 years of experience across a variety of sectors including DevOps, Automation, SDN, data center deployments, and even Multicast deployments across Service providers, he now spends his time tinkering with open source tools to make networking a lot more fun!
pdfTest your way to a better deployment!(PDF)
youtubeTest your way to a better deployment!
5:45pm - 6:00pm

Regency Ballroom (Banquet Level)

Closing Session

Meeting wrap-up, final thank you, future meeting announcements.

View full abstract page.
Speakers:

  • Betty Burke, NANOG
  • Currently serving as the NANOG Executive Director, responsible for all aspects of NANOG, reporting to the Board of Directors. Previous 37 years of experience serving in technology, business, and management within the Michigan Information Technology Services, University of Michigan, and Merit Network. Proven leadership and experience in development of strategic and operational plans, creation and implementation of marketing campaign for conference center and high tech facilities including a data center, conference and office building, library and campus fiber assets. Proven operational success through project management, along with leadership through community and team building.
youtubeClosing Session
pdfClosing Session(PDF)
6:00pm - 9:00pmInternational Ballroom (Lobby Level)SocialSponsors:

 

Thank you to the NANOG Service Sponsors:

Dyn, IMiller PR, ISC, Server Central

^ Back to Top