Heavy Networking

I see abstractions. I see automation. I see orchestration. I see people who think that because they can use a fancy management tool, they are an infrastructure expert. For networking, this drives me a little nuts. Technology fundamentals are the foundation of network engineering. By contrast, Ansible and Terraform are tools that help you build a network, but in no way replace those fundamentals. Sometimes I worry that tooling is replacing actual knowledge of how a network works. Abstractions–management layers, web UIs, magical scripts–may be replacing deep knowledge. Welcome to Heavy Networking, grumpy old network engineer edition. I’ve gathered a couple of other grumpy folks to chat about what young networkers should know but don’t. Our guests are Chris Young and Ivan Pepelnjak. We discuss: * The problems that come with a lack of understanding core concepts * What newer engineers do know * Just how deep core knowledge needs to go * Does Spanning Tree matter? * Are network engineers expected to be too multi-disciplinary? * More Sponsor: ThousandEyes ThousandEyes gives you performance visibility from every user to every app over any network, both internal and external, so you can migrate to the cloud, troubleshoot faster and deliver exceptional user experiences. Sign up for a free account at thousandeyes.com/packetpushers and choose a free ThousandEyes t-shirt. Sponsor: VIAVI Solutions VIAVI Solutions is a network performance management leader enabling IT teams to understand user experience with a patent-pending End-User Experience score. VIAVI delivers  outcomes, not just more data. Learn more at viavisolutions.com/packetpushers. Show Links: Chris Young on Twitter Kontrolissues – Chris Young’s site IP Space – Ivan Pelpeljak’s site Ivan Pepelnjak on Twitter

Is network automation just provisioning of equipment? That’s what most of us think of, but there’s more to the story. Anything within the scope of networking where we can reduce manual process falls under the umbrella of automation. Our sponsor today is NetBrain, and we’re going to discuss network automation from the standpoint of fixing problems–how a tool that deeply understands network topology can help you get to the bottom of a ticket without you having to query interfaces device by device while you troubleshoot. Joining us from NetBrain are Jason Baudreau, VP of Marketing and Chris Villemez, Sr. Technical Marketing Engineer. We discuss: * NetBrain’s roots in dynamic network mapping as an alternative to static diagrams * The role of dynamic mapping in automation, SDN, and cloud * How NetBrian builds a model of the network * Tying the network map to incident response to reduce Mean Time To Identify (MTTI) * How MTTI ties in to and speeds MTTR * Customer use cases * More Show Links: NetBrain

On today’s Heavy Networking episode we’re examining the state of SD-WAN. The technology is already being widely adopted to help companies cut WAN costs by trading MPLS for business broadband, and providing better visibility into and policy control over applications and performance in branch and remote offices. Where will SD-WAN go in the coming years? Will it swallow up branch security? How about end point and mobile device management? Could it extend its reach from the branch to become the way you manage your campus network? We’ll discuss all these questions and more. We cover: * The current state of SD-WAN * How SD-WAN affects cloud and security initiatives * The potential evolution of SD-WAN * Identity management and SD-WAN * More Sponsor: Cumulus Networks Cumulus Networks is announcing a brand-new networking certification: the Cumulus Certified Open Networking Professional. Learn Linux networking fundamentals, including essential concepts and commands behind Linux-based open networking, and master the Cumulus Core–everything you need to know to become proficient in Cumulus Linux. Get details at cumulusnetworks.com/cert. Sponsor: WAN Summit Networking professionals are gathering at the WAN Summit in New York, March 9-10. Join your network peers to understand how others are tackling the same challenges as you. Enterprise registrants get a special rate of only $99 to attend. Plus, you get an extra 20% off with the code PACKETPUSHERS when you register for New York. Get details at WANSummit.com. Show Links: A Unified Theory Of SDWAN Futures Part 1 – Greg Ferro via Ignition SDWAN: Get A Managed Service Or Do It Yourself? (Your Incompetence Is My Opportunity) – Greg Ferro via Ignition SD-WAN: Building The Business Case & Understanding Your ROI – Jason Gintert via Ignition SD-WAN Vendor List – Packet Pushers Heavy Networking 490: Lessons Learned From A Large SD-WAN Deployment – Packet Pushers Heavy Networking 483: SD-WAN Incompetence, Myths, And Fallacies – Packet Pushers

Cisco IOS XR version 7 is the topic of Heavy Networking in this sponsored episode. IOS XR7 is the latest version of the XR network OS (NOS). XR uses a modular architecture and is designed to support programmability, automation, and scale. It supports YANG models, real-time telemetry, APIs, and includes a robust routing stack. Cisco tells us XR7 is simple, modern, trustworthy, and cloud-enhanced. To get through to what that actually means, we’re chatting with Cisco guests Bhavna Prasad, Product Manager; and Reda Haddad, Distinguished Engineer. We discuss: * What’s new in version 7 * The ability to manage it with Ansible, Chef, or Puppet * Hardware support for IOS XR7, including whitebox * The API layer * Upgrading from older XR versions * New security features * More Show Links: Cisco IOS XR7 – Cisco Systems XR Docs An Architectural Approach to Flexible Consumption for Service Providers with IOS XR – Cisco Blogs Cisco Crosswork Network Automation – Cisco Systems Cisco Unveils Plan for Building Internet for the Next Decade of Digital Innovation – Cisco Systems End-of-Sale and End-of-Life Announcement for the Cisco IOS XR Software Release 6.6 – Cisco Systems IOS XR Coverage on Packet Pushers – Packet Pushers

A major automation challenge is where the information describing your network should live. You might think that your network would be best described by the config files on your switches and routers. Actually, that’s wrong. The configuration on the devices might–or might not–reflect what you intend the network to be. Plus, network device configurations aren’t easy to integrate with the rest of an IT provisioning process. What you require to facilitate network automation is a single source of truth. A source of truth that’s programmatically accessible, reflects intended state, and enables others to stand up infrastructure correctly without you getting in the middle of every provisioning request. Tim Schreyack joins us today to discuss a network automation solution using Ansible and Python, and of course, a single source of truth. Tim is a senior sales engineer at Dell Networking and has a background in network engineering and automation. He’s not here as a Dell spokesperson, and this isn’t a sponsored show. We discuss: * Why repeatability is the goal, and your organization’s size doesn’t matter * The pros and cons of using Ansible and Python as a singe source of truth * Jinja templates * Picking a database, and NetBox pros and cons * Automation differences in greenfields and brownfields * Where to start with automation * More Sponsor: Viavi Solutions VIAVI Solutions is a network performance management leader enabling IT teams to understand user experience so they can solve performance problems fast. Learn more at viavisolutions.com/packetpushers. Sponsor: ThousandEyes ThousandEyes gives you performance visibility from every user to every app over any network, both internal and external, so you can migrate to the cloud, troubleshoot faster and deliver exceptional user experiences. Sign up for a free account at thousandeyes.com/packetpushers and choose a free ThousandEyes t-shirt. Show Links: Powering Your Automation: A Single Source of Truth – CHINOG – June 27, 2019 – YouTube Powering Your Automation: A Single Source of Truth – NANOG – November 5, 2019 – YouTube Tim Schreyack on LinkedIn Heavy Networking 452: Using NetBox As A Source Of Networking Truth – Packet Pushers

If one of your New Year’s resolutions is to blog more, or start a blog, this episode is for you. Technical blogging comes with benefits. For one, writing about a subject helps you understand it better (and also helps you realize the limits of your knowledge). Blogging can raise your profile. When you share information or experiences with a larger audience, you can develop a reputation for expertise. This higher profile may create  opportunities to speak at events, get access to conferences or early product releases, and maybe even get a better job. A technical blog can serve as a kind of living resume that potential employers and recruiters can see to get a sense of your capabilities. When you share technical information, troubleshooting tips, study hacks, analysis, and other content, you help other IT pros who may be looking for information. It’s also not uncommon that you’ll help yourself–more than one blogger has run into a problem, hit up Google, and found an old post with a solution. Our guests for today’s show are John Mark Troyer, founder of the Influencer Marketing Council and TechReckoning; and Stephen Foskett, founder of Tech Field Day and GestaltIT. We discuss: * The motivations and purposes for blogging * How to get started * Favorite blogging tools and writing tips * The “small b” approach to blogging (don’t chase metrics, not every post has to be epic, etc.) * Whether to mix personal and professional writing * More Sponsor: Cradlepoint The future of the WAN is wireless, but a wireless connection is only as good as the edge. Cradlepoint unlocks the power of advanced cellular through wireless edge solutions that are delivered the way you consume everything IT: as a service. Reliable. Elastic. Simple to manage from anywhere. Learn more about Cradlepoint’s cloud-managed LTE solutions at cradlepoint.com/packetpushers. Sponsor: ThousandEyes ThousandEyes gives you performance visibility from every user to every app over any network, both internal and external, so you can migrate to the cloud, troubleshoot faster and deliver exceptional user experiences. Sign up for a free account at thousandeyes.com/packetpushers and choose a free ThousandEyes t-shirt. Show Links: Tech Field Day GestaltIT.com Influencer Marketing Council Techreckoning.com Stephen Foskett on Twitter Jon Mark Troyer on Twitter Arse First Method of Technical Blogging [PDF/iPad/Kindle] – Greg Ferro

On December 18, 2019, the Packet Pushers hosted a livestream gathering on YouTube where the Packet Pushers and special guests answered audience questions. This podcast episode is the audio capture of that livestream. Our conversation covers: * Cisco’s new ASIC and 8000 router family * Will the multi-cloud ever get a hypervisor, and what role will Kubernetes play? * Cisco’s new certification path and whether certs teach deep knowledge or specialized product operation * Will SDA and ACI merge into a single fabric? * The current state of VNF performance * Zero-trust environments and software-defined perimeters This is our final Heavy Networking show for 2019. Thanks for another year of your attention, support, and conversation. We’ll be back in January 2020 with more nerdy networking and IT shows to keep you company in the car, at the gym, or wherever you listen. Enjoy the holidays and have a Happy New Year! Sponsor: ITProTV Get over in-depth technical training from ITProTV. ITProTV offers online instruction in CompTIA, Cisco, VMWare, Microsoft and more. You can stream courses live and on demand on your favorite device. Sign up at itpro.tv/packet and save 25%. Use the code PACKET25 when you check out. Sponsor: ThousandEyes ThousandEyes gives you performance visibility from every user to every app over any network, both internal and external, so you can migrate to the cloud, troubleshoot faster and deliver exceptional user experiences. Sign up for a free account at thousandeyes.com/packetpushers and choose a free ThousandEyes t-shirt. Show Links: Ethan Banks Greg Ferro Tommy McNicholas Ned Bellavance Drew Conry-Murray Russ White Tom Bragg Network Break 265: Cisco Chips At Broadcom With New ASIC; AWS Gives Networking Some Love – Packet Pushers Day Two Cloud 027: Do Enterprises Need Kubernetes? – Packet Pushers The Hedge Episode 5: Geoff Huston on DoH – The Hedge The Hedge Episode 6: Geoff Huston on DoH – The Hedge Packet Pushers Holiday Edition Livestream December 18, 2019 – The Packet Pushers

Cloud and digital transformation initiatives are pushing security operations teams to the limit. The growing complexity of cloud and on-premises environments make it harder for security operations to keep pace with developers, manage proper access to resources, and ensure that the right controls are in place and policies are met. It’s difficult for security teams to implement approaches such as zero trust and microsegmentation because they are swamped with day-to-day tasks. At the same time, organizations can’t just throw more bodies at the problem because it’s hard to recruit and train security professionals. On today’s sponsored podcast we examine how automation can help improve the effectiveness of security operations. Joining us is Ofer Or, Vice President of Products at Tufin, our sponsor for this episode. We discuss: * The growing complexity of enterprise environments on premises and in the cloud * Automating labor-intensive tasks to improve the effectiveness of security operations * Getting proper visualization of your infrastructure to understand policy impact * Benefits of network segmentation on premises and in the cloud, and how to implement it * How Tufin products such as Iris and Orca address policy management and zero trust security for cloud applications * More Show Links: Tufin.com  Tufin Iris Tufin Orca

Welcome to Heavy Networking. Joining us is William Collins, lead cloud architect for a large healthcare company, who raised his hand when we asked for folks who wanted to be a guest on the show to volunteer. William volunteered, and we’re going to discuss hybrid cloud networking. Specifically, we’re going to drill into routing complexity. No one operating public cloud networks had “making the network engineer’s life easier” at the top of their list when they came up with things like ExpressRoute and VPCs. William will share his practical and hard-earned experiences connecting applications and services from on premises into the public cloud. We discuss: * A working definition of hybrid cloud * Developing a hybrid cloud strategy * The importance of design and governance * Cloud connectivity pros and cons for VPNs, DirectConnect, ExpressRoute, and Carrier Neutral Facilities (CNFs) * Routing topology scenarios * Why your networking knowledge is relevant–and crucial–when it comes to the cloud * More Sponsor: INE This episode is sponsored in part by INE, the ‘experts at making you an expert’. INE is announcing a new, monthly All Access Pass subscription plan, which provides you with unlimited access to INE’s entire content library of over 14,000 of the best IT and Networking training videos for just $99 a month. Visit INE.com/packetpushers to get started. Sponsor: ExtraHop Today’s show is also sponsored in part by ExtraHop, the enterprise cyber analytics company delivering performance and security from the inside out. ExtraHop offers complete visibility with machine learning to help you make quick, confident decisions about your IT environment. Explore the ExtraHop Performance Platform at extrahop.com/packetpushers. Show Links: William Collins on LinkedIn William Collins on Twitter Heavy Networking 452 – Using Netbox As A Source Of Networking Truth Cloud Certifications AWS Certification Microsoft Azure training Google Cloud certified Cloud Documentation AWS Documentation Microsoft Azure Documentation Google Cloud Documentation

With 5G, we have a new set of challenges for service provider networks. As the network becomes increasingly dynamic and distributed to deliver an ever-evolving set of services, providers have to contend with increased complexity. To explain this complexity and what can be done to wrangle it, sponsor Juniper Networks has sent Amit Bhardwaj, Director of Product Management, to the palatial Packet Pushers virtual studio. We discuss: * The potential business opportunities of 5G for service providers * 5G use cases such as IoT, automotive, smart cities, gaming, and others * Network slicing vs. L3 VPNs * The need to secure devices and the network at scale * Why automation is critical to deliver new services and capabilities * Contrail HealthBot * More Show Links: Juniper Networks Juniper Networks Service Provider Solutions Juniper Networks NorthStar Controller Juniper Contrail HealthBot Heavy Networking 477: Segment Routing Boot Camp With Juniper Networks (Sponsored) – Packet Pushers Amit Bhardwaj on LinkedIn Amit Bhardwaj on Twitter

Here are some things your SNMP-polling based network monitoring station probably doesn’t do: 1. Consume inbound streaming telemetry 2. Comprehend YANG models 3. Analyze network events and percolate up the things you need to address This matters because network monitoring is moving from graphing what happened during the last polling cycle to interpreting network events in real time and letting you know what to do about it. Joining us from Juniper is Javier Antich. Javier is Senior PLM Director in the Automation Software team at Juniper, with a focus on automation, programmability, telemetry, and analytics. Javier and the Packet Pushers discuss: * What’s driving automation in the enterprise and why it matters * The differences between telemetry and traditional monitoring data * How to make telemetry consumable and relevant * The role of AI and machine learning * An overview of Juniper’s HealthBot automation tool * More Show Links: Contrail HealthBot – Juniper Networks Automated Network Operations – Juniper Networks

Segment routing is a mechanism for sending packets over a specific path through the network. There are several ways to do segment routing and multiple elements that go into the technology. On today’s Heavy Networking podcast we explore segment routing as it’s used in a production network; specifically, ESnet, an international network that moves science data all over the world, including data from the Large Hadron Collider in Europe to scientists in the United States. Our guest is Nick Buraglio, a network engineer at ESnet and a regular guest on Packet Pushers. We discuss: * The flavor of segment routing that ESnet went with and why (SR-MPLS vs. SRv6) * The problems ESnet was trying to solve * Whether segment routing reduces complexity * The role of IS-IS in the network * Path Computation Element (PCE) and how forwarding decisions are made * More Sponsor: Cradlepoint The future of the WAN is wireless, but a wireless connection is only as good as the edge. Cradlepoint unlocks the power of advanced cellular through wireless edge solutions that are delivered the way you consume everything IT: as a service. Reliable. Elastic. Simple to manage from anywhere. Learn more about Cradlepoint’s cloud-managed LTE solutions at cradlepoint.com/packetpushers. Sponsor: Cumulus Networks Today’s show is sponsored by Cumulus Networks. Build a private cloud environment that is open, modern, and simpler to manage with Cumulus Linux – the Linux-based open network OS that gives you total interoperability, and NetQ – which gives you end-to-end actionable insight from the host to the switch. To learn more about open networking, head to https://cumulusnetworks.com/open. Show Links: Heavy Networking 477: Segment Routing Boot Camp With Juniper Networks (Sponsored) – Packet Pushers A Glimpse At Two Approaches To Segment Routing – Ethan Banks Show 370: Cisco & IPv6 Segment Routing (Sponsored) – Packet Pushers Yet another blog about Segment Routing-Part 1 – Diptanshu Singh via Packet Pushers Yet another blog about Segment Routing-Part 2 : TI-LFA – Diptanshu Singh via Packet Pushers Yet another blog about Segment Routing-Part 3: SR-TE – Diptanshu Singh via Packet Pushers Nick Buraglio on Twitter Forwardingplane.net – Nick Buraglio’s blog

Heavy Networking visits with Snehal Patel. A global network architect for a brand-name retailer, Shehal has been deploying SD-WAN since before it became the biggest buzzword to bet your business’s behind on. With one of the largest active enterprise SD-WAN deployments in the world, we asked him if he’d be willing to come on the show and let us know how it’s going. How large? More than 2,800 retail locations in North America alone, all live and in production on an SD-WAN fabric. His WAN connections are a mix of Internet broadband, LTE, and MPLS. Snehal joins us to discuss his experiences, both good and bad, with deploying and running SD-WAN at scale. We explore: * The general environment he’s rolling out SD-WAN for * How branches connect to HQ and cloud services * Security and compliance requirements that have to be met * How ZTP isn’t quite Zero Touch, but still an improvement * The cutover process and how it’s been refined * The good and bad of centralized management * Much more Sponsor: ITProTV Get over in-depth technical training from ITProTV. ITProTV offers online instruction in CompTIA, Cisco, VMWare, Microsoft and more. You can stream courses live and on demand on your favorite device. Sign up at itpro.tv/packet and save 25%. Use the code PACKET25 when you check out. Sponsor: ExtraHop ExtraHop is the enterprise cyber analytics company delivering performance and security from the inside out. ExtraHop offers complete visibility with machine learning to help you make quick, confident decisions about your IT environment. Explore the ExtraHop Performance Platform at extrahop.com/packetpushers. Show Links: Snehal Patel on Twitter ProSDN – Snehal’s blog Snehal Patel on LinkedIn

BBR is a congestion control algorithm (CCA) designed for low latency. Backed by Google, BBR’s use is growing on the Internet, including as the default CCA for YouTube. However, an element in BBR’s design means that the 1.0 version of the algorithm crowds out other commonly used CCAs, such as Cubic and Reno. As today’s guests discovered in their research, when BBR and Cubic flows share a link, a single BBR flow will take up as much as 40% of the link’s bandwidth, leaving 16 Cubic flows to divide the rest among themselves. This leads to questions about how BBR will interact with the legacy algorithms that are the status-quo today. Our guests for today’s show are Ranysha Ware, a Ph.D. student in Computer Science at Carnegie Mellon University; and Justine Sherry, Assistant Professor of Computer Science at Carnegie Mellon. Ranysha recently presented her findings on BBR at the Internet Measurement Conference in October 2019 in the Netherlands. Ranysha and Justine come on the podcast to discuss: * How BBR works, including differences with legacy loss-based CCAs * Research comparing BBR, Cubic, and Reno * Why BBR behaves the way it does * Potential impacts for the Internet * Using harm, rather than fairness, as a measure for new algorithms * More Ranysha has also created a Congestion Control Evaluation Survey for network engineers and operators to get feedback on a proposed testbed for evaluating the impact of CCAs on applications and networks. She’d appreciate your input! Sponsor: Tufin Tufin has pioneered a policy-based approach to network security management using automation and analytics. As a result, you can make network changes in minutes instead of days, reliably and securely. Tufin. The Security Policy Company. Get details at www.tufin.com. Sponsor: ThousandEyes ThousandEyes gives you performance visibility from every user to every app over any network, both internal and external, so you can migrate to the cloud, troubleshoot faster and deliver exceptional user experiences. Sign up for a free account at thousandeyes.com/packetpushers and choose a free ThousandEyes t-shirt. Show Links: Modeling BBR’s Interactions withLoss-Based Congestion Control (PDF) Ranysha Ware – Carnegie Mellon University Congestion Control Evaluation Survey Ranysha Ware on Twitter Dr. Justine Sherry Justine Sherry on Twitter BBR Congestion Control Work at Google – IETF Employing QUIC Protocol to Optimize Uber’s App Performance – Uber Engineering Blog

Today on Heavy Networking, we’re going to evolve using genetic algorithms. You heard me right. Researchers at the University of Maryland in the United States have developed a project called Geneva that uses genetic algorithms to automatically figure out the best way to, in this use case, avoid Internet censorship. Essentially, we’re going to discuss how to get through the Great Firewall of China with about 6,000 lines of Python. So put on your propellor beanie, and let’s introduce our guests: David Levin, Assistant Professor, Computer Science and Co-Chair, CS Undergraduate Honors Program at the University of Maryland; and Kevin Bock, a graduate student at the University of Maryland who did much of the work on this project. We discuss: * Common techniques state censors use to detect and stop traffic, including encrypted traffic * Limits to these approaches that Geneva can exploit * How genetic algorithms work * Training genetic algorithms against censorship techniques * Testing mechanisms, including against real-world censors * Could Geneva be used to thwart legitimate security controls * More Sponsor: ITProTV Get over in-depth technical training from ITProTV. ITProTV offers online instruction in CompTIA, Cisco, VMWare, Microsoft and more. You can stream courses live and on demand on your favorite device. Sign up at itpro.tv/packet and save 25%. Use the code PACKET25 when you check out. Sponsor: ExtraHop ExtraHop is the enterprise cyber analytics company delivering performance and security from the inside out. ExtraHop offers complete visibility with machine learning to help you make quick, confident decisions about your IT environment. Explore the ExtraHop Performance Platform at extrahop.com/packetpushers. Show Links: Geneva: Evolving Censorship Evasion – University of Maryland Geneva – GitHub Dave Levin on Twitter PQ 58: Alibi Routing With UMD’s Dave Levin – Packet Pushers