Heavy Networking

Today on Heavy Networking, a career-oriented show. We’re discussing how to take those complex engineering systems in your brain and share them in a way other people can understand them. Maybe that’s a co-worker. Maybe that’s your boss. Maybe that’s a presentation to your customer who needs to make a buying decision. Maybe that’s a presentation to the entire company because you’re training them on a new VPN login procedure. Our guest today is Drew Conry-Murray. Yes, *that* Drew, the voice you hear every week on Tech Bytes and Network Break, and guest hosting on other podcasts across the Packet Pushers podcast network. I’ve known Drew for many years. He was my editor when I wrote for Network Computing and Information Week. He was also my handler when I was delivering presentations at Interop conferences. And of course in recent years, Drew has become a Packet Pusher, slaving away in the packet mines alongside me and Greg Ferro. We discuss: * The challenges of communicating technical ideas and concepts * Balancing technical detail with broader outcomes * Why assuming non-technical people are stupid is a bad idea * The power of editing * The value of feedback * Helpful writing tips * More Sponsor: CBT Nuggets CBT Nuggets is IT training for IT professionals and anyone looking to build IT skills. If you want to make fully operational your networking, security, cloud, automation, or DevOps battle station visit cbtnuggets.com/heavynetworking. Show Links: How To Be A Better Tech Writer – Human Infrastructure Newsletter We Need A Better Name For ‘Soft Skills’ – Packet Pushers Document So You Can Go On Vacation – Packet Pushers Overcoming The 5 Objections To Tech Blogging – Packet Pushers

Most people consider ‘cloud’ to be limited to a handful of hyperscale giants, but the fact is any company can build cloud infrastructure. It doesn’t have to be in massive data centers in dozens of locations around the world. Juniper Networks is here in this sponsored episode to make the case that service providers are building cloud services too, and that enterprises should consider these options. Joining us is Kevin Hutchins, SVP, Strategy & Product Management at Juniper Networks, to assert that service providers can thrive and compete in a cloud-based economy. He also argues that Juniper will be a key part of the ecosystem. “Cloud at the Edge” if you will. We discuss: * Why service providers are expanding from connectivity to digital services * How service providers can leverage their real estate and experience in managing distributed systems * 5G as a driver * The rise of the edge cloud * Juniper’s role in enabling service providers to operate cloud services * Enterprise use cases for a service provider edge cloud * More Show Links: Telco Cloud Transformation in the 5G Era – Juniper Networks Telco Cloud: Your Platform for Agile Service Delivery – Juniper Networks Engineering Simplicity: Delivering a Future Distributed Edge Cloud Today – Juniper Networks Keep Your Edge – Monetizing Without Losing Control – Juniper Networks StackPath Edge Computing Platform Receives Investment from Juniper Networks & Cox Communications – Juniper Networks Juniper CTO Dishes Edge Cloud Strategy – Juniper Networks Juniper Networks exec: How carriers can build a better telco cloud without large cloud providers – Juniper Networks Juniper CTO welcomes Nokia to data center switching arena – Juniper Networks 2020 Next-Gen Edge: The Role Operators can play at the Edge. – Juniper Networks

Intel has a substantial position in networking silicon and technologies to accelerate network functions. You might be familiar with the chip maker’s 2019 acquisition of Barefoot Networks. This acquisition brought Barefoot’s programmable switch ASIC, Tofino, into Intel’s stable. Intel is also a significant contributor to the P4 language for programming ASICs. There’s also Intel’s investment in DPDK, which optimizes x86 CPUs for network use cases; plus a product line that spans NICs, SmartNICs, FPGAs, and more. On today’s podcast, sponsored by Intel, we dive into Intel’s portfolio to understand how it unlocks the compute power of your data center. Our guest is Mike Zeile, Data Center Group Vice President and General Manager of End-to-End Network Applications at Intel. We discuss: * The widespread adoption of the P4 language * Extending P4 from switches to SmartNICs, appliances, and software pipelines * Intel’s progression with the Tofio ASIC * Leveraging SmartNICs for acceleration * Using eBPF and XDP for integrated telemetry * A glimpse into the future of silicon photonics * More Show Links: Removing Roadblocks in Data Movement – Intel Better Connectivity Means Better Experiences – Intel

Does EIGRP need defending? Some of you might argue that there is no defense for this protocol. No matter which side of the divide you fall on, we’re having the EIGRP discussion. We’re going to go through major use cases, design considerations, scaling tips, and more. Our guest is Zig Zsiga. You might know him as the voice of Zig’s network design podcast where he discusses how to think about network design. Zig and I know each other from the internet, and you might remember him from episode 555, where we discussed the top network design trends of 2020. We explore: * Why EIGRP has a bad reputation * Whether networks are still using EIGRP * Major use cases and business requirements * Design recommendations * More Sponsor: CBT Nuggets CBT Nuggets is IT training for IT professionals and anyone looking to build IT skills. If you want to make fully operational your networking, security, cloud, automation, or DevOps battle station visit cbtnuggets.com/heavynetworking. Show Links: Heavy Networking 555: Top Network Design Trends Of 2020 – Packet Pushers Zigbits Network Design Podcast Zigbits Network Design Videos – YouTube Zigbits Courses Zigbits Articles/Blogs @zig_zsiga – Zig on Twitter Zig on LinkedIn

Today’s Heavy Networking episode is a grab bag of topics delivered in our community roundtable format. We’ve brought in folks from our public Slack channel who wanted to talk about networking stuff that’s on their mind. (Go to packetpushers.net.slack, read the rules, and sign up if you want to join the group). Today’s grab bag of ideas will lay bare for all to see the soul of networkers the world over, and we’re gonna talk through these oversized packets with no fragmentation allowed and the evil bit on, crammed right into your ears, and you can’t stop us. It’s happening. It’s happening right now. Joining us around the community round table are Justin Seabrook-Rocha, a Network Engineering Manager; Ken Celenza, the Managing Director at NetworkToCode; Nick Buraglio, a Planning and Architecture Engineer; Michael Schipp, a Pre-sales Engineer; and Steve Puluka, an IP Architect at DQE Communications. We’ll see how far we get in our discussion today, but topics up on the virtual whiteboard include: * IPv6 only * Firewall rule lifecycle management * Keeping up as a manager * Evaluating tech that’s being hyped * Whitebox switching * Smart NICs Show Links: @xenith – Justin Seabrook-Rocha on Twitter @itdependsnet – Ken Celenza on Twitter @forwardingplane – Nick Buraglio on Twitter ForwardingPlane.net – Nick’s blog @maschipp – Michael Schipp on Twitter @spuluka – Steve Puluka on Twitter Puluka.com – Steve’s blog Packet Pushers Slack Channel

Welcome to Packet Pushers Heavy Networking. We are deep in the software defined network transition in 2021 and, as an industry, learning more about software operated networks. In the first phase, SDN focused on making configuration changes using software. The early adopters learned that configuration at scale needed assurance and verification. What do I mean? You can’t configure a network without awareness of the entire system. In simplistic terms, you can’t configure a simple VLAN without doing assurance check for STP, available IDs, and so on. In today’s sponsored podcast with IP Fabric, you will get an appreciation of the company’s vision of using intent based networking for verification and assurance. Remember that intent simplifies to “results” and leans into software to assist with getting those results. The IP Fabric Network Assurance platform gathers network configuration and state, and then builds a network model using a graph database. The result is a “digital twin” of the network that engineers and security teams can use for troubleshooting, to plan and verify changes, for network and security analysis, and to enable closed-loop automation. Our guests from IP Fabric are Pavel Bykov, CEO; and Daren Fulwell, Network Automation Evangelist. We discuss: * How IP Fabric builds its model by collecting state and configuration data * Using the model to map and test changes, enhance troubleshooting, and get visibility * Getting from intent to actual network outcomes * How IP Fabric can enable closed-loop automation * Customer case studies * More Show Links: IPFabric.io @IPFabric – IP Fabric on Twitter IPFabric on LinkedIn

What do you do when you’ve got a big ol’ pile of data that you need to process, but it’s out in the middle of digital nowhere? You could build an edge network and computing stack, perhaps. That’s all the rage for some industries–LOTS of digital ink spilled and consortia formed about edge computing. But what if an edge solution is not an option for you? What if you really need to get that big ol’ data pile up to the cloud or an HQ site for processing? You need to come up with some unusual networking solutions to get the data from where it is to where it needs to go. Our guest today is Michael Shepherd, and he’s had to address this exact problem, getting video from sites where there isn’t any infrastructure up to public cloud and other places for processing. Michael is CTO at Rogers-O’Brien Construction. We discuss: * Addressing a lack of L1 connectivity at work sites * Construction is a data-driven business with 4K video, site surveys, 3D modeling, augmented reality, and more * Using drones and robot dogs to collect site video * Bandwidth and resilience issues with LTE and broadband * Getting data from the work site to the HQ and cloud * The operational impacts of SD-WAN * More Show Links: Michael Shepherd on LinkedIn Tech Bytes: Construction Firm Improves Job Site Productivity With Silver Peak SD-WAN (Sponsored) – Packet Pushers

On today’s sponsored Heavy Networking we talk to Nokia Networks about SR Linux, a new network OS. SR Linux is built on a microservices architecture with modern data center networking features. It’s designed to fit a variety of use cases while also being automation-friendly and extensible. Our guest is Bruce Wallis, Senior Director of Product Management in Data Center Switching at Nokia. We discuss: * Building a NOS from a clean slate * Serving turn-key use cases while also enabling customization and extensibility * How and why SR Linux leverages protobuffs, gRPC, and gNMI * Enabling NetDevOps via programmatic interfaces * Making telemetry more consumable * More Show Links: SR Linux Tech Video – Nokia SR Linux Product Description – Nokia Nokia Service Router Linux (SR Linux) – Nokia Nokia Data Center Fabric – Nokia Data Center tour launch video – Nokia Fabric Services Platform Product Description – Nokia Nokia Hardware Platforms For Data Center Fabrics – Nokia Heavy Networking 538: Running An Open, Automated Data Center Fabric With Nokia (Sponsored) – Packet Pushers

Network Functions Virtualization (NFV). Taking that thing that used to be a router or firewall or load balancer on fancy hardware and sticking it on a generic x86 or ARM server. You now have a virtualized version of the network function. Pretty handy when you’re constrained for rack space or want to spin these things up and down on demand. Cloud friendly. Multi-tenant friendly. Service chaining friendly. But if you love to hug your router, maybe you’d rather avoid NFV. Maybe you’d rather run physical networking hardware with very special silicon and virtualize that. And you have good reasons. And stories. And scars. It’s 2021. Are your reasons for hating on NFV still good reasons? Our guests today think just the opposite. In fact, they don’t have time for hardware. Let’s talk through what that means with Michael Pfeiffer, Cloud Networking Architect working in the salt mines at a VAR; and Brad Gregory, Senior Product Manager at Equinix. This isn’t a sponsored show. We’re just going down the road of seeing if the NFV juice is worth the squeeze these days. We discuss: * Whether NFV performance on commodity x86 is sufficient * Overcoming trust and control issues with NFV providers * Whether NFV instances are ephemeral or forever * How much operational effort is required to adopt NFV * More Show Notes: Michael Pfeiffer on LinkedIn Brad Gregory on LinkedIn Ethan Banks on Twitter Greg Ferro on Twitter

On today’s Heavy Networking we talk about experience monitoring, the whole idea of which is to get a sense of what the user perceives when they consume an application. IT folks tend to view this issue from their own particular area of responsibility–networking, security, app development–but the reality is there’s a common set of data that IT should look at. The industry calls this Digital Experience Monitoring (DEM). DEM borrows from network monitoring, but enhances that information with other sources (synthetic transactions, Internet performance metrics, page load times, and so on). Catchpoint, our sponsor for this episode, provides DEM tools for monitoring the end user experience. Our guest is JP Blaho, Director, Product Marketing at Catchpoint. We discuss: * How Catchpoint combines user and Internet data * How Catchpoint differs from traditional network monitoring tools * The need to share relevant information across IT silos * The role of the Site Reliability Engineer (SRE), and whether your organization needs one * What experience monitoring looks like in a post-pandemic environment * More Show Links: Catchpoint Network Insights @catchpoint – Catchpoint on Twitter Heavy Networking 547: Building And Monitoring A User-Centric Digital Experience With Catchpoint (Sponsored) – Packet Pushers @jpblaho – JP Blaho on Twitter

GNS3 is a tool that folks use to create virtual networks for labbing purposes. The environment allows you to run binaries under QEMU, Docker container images, and virtual machines, plumbing them together using an intuitive UI. GNS3 is a sort of magic, and it’s free. To be clear, you have to supply your own networking device images to run in the GNS3 environment. GNS3 is just that–an environment to run device images you have the legal rights to run. But GNS3 is the environment that makes it easy to plumb those images together and see what they can do. There are certainly other tools in this space both open source and commercial, but GNS3 happens to be my go-to for lab work. In this episode, Heavy Networking welcomes GNS3 co-founder and developer Jeremy Grossman and networking instructor David Bombal. We’re going to cover the state of GNS3 in 2021, including what GNS3 can do that maybe you didn’t know, and what’s on the roadmap. If the last thing you did in GNS3 was calculate an idle PC value, there are a ton of new capabilities for you to catch up on. Show Links: GNS3 GNS3 Docs GNS3 Community GNS3 Academy David Bombal’s YouTube channel GNS3 Docker Persistent Containers – YouTube Submitting a new appliance template – GNS3.com David Bombal on Twitter

Today’s episode is the last Heavy Networking of 2020. In the spirit of an end-of-year reflection, we’re going to talk about network design trends from this year, some of which were driven, or at least accelerated, by the pandemic hellscape that was 2020. Our guest is Zig Zsiga, a network designer, architect, and CCDE who also does training and education at zigbits.tech. See the show links below to find out all the ways you can connect with and learn from Zig. We discuss: * Having to support remote work/distributed work at scale very quickly * Changing governance and security models to support remote work * The accelerating growth of workloads in the public cloud * New operational models around automation and DevOps * Data center and campus fabrics * Zero Trust Access * The rise of end user experience as a performance metric * More Sponsor: InterOptic InterOptic makes high quality optical modules you can rely on. Plus, they are far cheaper than OEM optics. Save big money without compromising quality. Visit interoptic.com/packet-pushers. Sponsor: DriveNets DriveNets offers a network cloud: a hugely scalable network on a disaggregated platform. Get higher capacity at a lower cost. Find out more at drivenets.com/resources. Show Links: Zigbits Network Design Podcast Zigbits Network Design Videos – YouTube Zigbits Courses Zigbits Articles/Blogs @zig_zsiga – Zig on Twitter Zig on LinkedIn

In November 2020, I wrote an article for our Human Infrastructure newsletter where I compared open source software to adopting a puppy. Puppies are cute and fun, but you have to train them to be a dog you can live with. I further compared commercial offerings to being more like a grown-up dog that’s already been trained for you. My point was that there’s a tradeoff between open source and commercial software. You pay in money for commercial software. But that doesn’t make open source software free. With OSS, you pay a DIY tax to make it do all the things you might need it to do. Our guest today is Daniel Teycheney. Daniel tweeted at me about this article, arguing that commercial software can often be a puppy, too. So we’re gonna talk about it. Is there really a tradeoff in OSS vs. commercial software? Has commercial software gotten so bad that you might as well go open source? Is open source not as time intensive as I’m making it out to be? Let the debate begin! Show Links: WriteMeme – Daniel Teycheney on GitHub Motherstarter – A network data transformation tool to accelerate the adoption of network automation by providing a command-line interface to convert input file sources to various network automation output formats. – GitHub DanielTeycheney.com Daniel Teycheney on LinkedIn @danielteycheney – Daniel Teycheney on Twitter Open Source Puppies Need Lots Of Attention – Packet Pushers newsletter

BGP link-state is not new. It’s been kicking around since 2015 or so. On the surface, you might think, “Why are we carrying around link state information in BGP? That’s not what BGP is for.” You’re not the only one with that reaction. But BGP-LS is more nuanced. Engineering choices are made for a reason, and remember that everything’s a trade off. To educate us on what BGP-LS is and why it exists, we welcome Hannes Gredler. He’s heavily involved in the deepest corners of the networking industry as the CTO and founder of startup RtBrick, offering cloud-native routing software for telcos. He’s also chaired the IETF IS-IS working group, and was a Distinguished Engineer at Juniper Networks. He holds some routing patents. He’s a published author. Hannes knows what’s going on when it comes to BGP-LS, and volunteered to share some of his time with Heavy Networking. We discuss: * What BGP-LS is and the problems it solves * How BGP-LS was conceived * Why carry link state in BGP? * Are we stuffing too many features into BGP? * How BGP-LS compares to segment routing * More Sponsor: ITProTV Start or grow your IT career with online IT training from ITProTV. There’s a special offer for Packet Pushers Heavy Networking listeners: sign up and save 30% off all plans! Visit itpro.tv/PACKETPUSHERS for 30% off all plans. Use promo code PACKETPUSHERS at checkout. Show Links: @HannesGredler – Hannes Gredler on Twitter Hannes Gredler on LinkedIn The Complete IS-IS Routing Protocol – Hannes Gredler’s book, available on Amazon RtBrick blog – RtBrick Heavy Networking 477: Segment Routing Boot Camp With Juniper Networks (Sponsored) – Packet Pushers

On today’s Heavy Networking, sponsored by Fortinet, we dive into a variety of topics including a multiplicity of edges that require connectivity and security (WAN edge, LAN edge, cloud edge, remote edge, etc.), the convergence of infrastructure and security, the need for interoperability among security vendors, implementing Zero Trust Access (ZTA), and the current and future roles for technologies such as machine learning (ML) and AI. Our guest today is Fortinet’s John Maddison, EVP of Products. John shares his insights on why customers are converging on fewer vendor platforms, how security features and capabilities are consolidating, and how Fortinet integrates ML and AI. We discuss: * Protecting edge environments * Why customers are converging on fewer platforms * The consolidation of security functions * Enabling Zero Trust Access (ZTA) * Fortinet’s use of connectors to integrates with third-party products * How Fortinet uses machine learning to improve security operations and response * The company’s AI roadmap * More Show Links: Network Security Expert (NSE) Institute – Fortinet Fortinet.com Free Cybersecurity Training