DENOG13

Welcome to our workshop day

We will be going through
* Grafana 101
* How to get data into Prometheus and Loki
* How to get that data into Grafana
* Some best practices of networking dashboards
* How to alert on it all

Introduction to BGP workshop in 5 hours.

Topics:
- Prefixes and Autonomous Systems
- Introduction to BGP
- DE-CIX Academy BGP Lab
- IGP and iBGP
- eBGP
- Multihoming
- BGP Best Path Selection

Participants should have a setup with two screens or one large screen.

Work days of network operators can become difficult - especially the moment when you have to involve another party like the vendor of your network equipment. In this workshop
we will not only give you the view from a TAC engineer working on your cases, but also examples on how a TAC actually works and what will be needed to make it a pleasent experience for both sides.

Together we will have a look at some real world examples of Do's and Don't when it comes engaging with TAC.

From 0 to a full blown EVPN/VXLAN leaf-spine datacenter deployment in <2h? Sounds like a fairy tale.
In this workshop Patrick will show you interactively (participants get there own enviornment) how to define and configure a datacenter design in YAML and Ansible based on the Arista Validated Deisgn. It includes a set of ansible roles and modules to help kick-start your automation.

Linux ist seit einigen Jahren ein 1st Class Networking Citizen und muss sich schon lange nicht mehr
hinter kommerziellen Lösungen verstecken. Es kommt gar in vielen solchen zum Einsatz und bildet das
Fundament für fast alle Cloud-Lösungen da draußen.

In diesem Workshop wird es um Methoden und Features zur Layer3 Netzwerk-Separierung gehen und wir werden uns

• Policy-based Routing
• VRFs (mit und ohne MPLS)
• Network Namespaces

anschauen, die Möglichkeiten vergleichen und Einsatzmöglichkeiten durchgehen. Dabei wird es dann
u.a. um Linux als Router, VPN-Server, Load-Balancer, etc. gehen. Fragen und Ideen sind gewünscht und dürfen gerne vorher bei max@sdn.clinic eingeworfen werden. Es wird ein kleines Lab geben in dem wir ggf. live Dinge ausprobieren können. Bringt Zeit, Tee und Kekse mit!

This workshop will cover basics of segment routing along with live demo of configuring segment routing and migrating the network to start using segment routing. The workshop will then go deeper into segment routing and cover traffic-engineering with flex-algo and inter-domain traffic-engineering with BGP-CT.

Join to learn about the multiple aspects of DENOG

Presentation (no lab, sorry) about the basic stuff of networking:

• Packets and Protocols
• Ethernet
• VLANs
• IP
• IP Addresses and Routing
• Global IP routing
• UDP
• TCP

Welcome to DENOG13!

BNG Blaster - https://github.com/rtbrick/bngblaster

When we started to build and test our BNG software for open hardware platforms, we got increasingly frustrated with the abilities and test methodology of commercial network testing software. Most solutions come with a dedicated hardware chassis or VMs with high CPU and memory requirements. Commercial network testing software is often limited by strict licensing. When performing regression testing overnight on cloud servers we have a peak license use of hundreds of instances, which we do not need most of the day. Our reliability testing crew complains often about slow reconfiguration from one test scenario to the next. Reprogramming the network tester often takes an order of minutes before the test suite can be run but we want to start/stop a test within an order of seconds.

Therefore we decided to build an open source network test software initially focused on BNG and IPTV testing but constantly enhanced and planned for more common network equipment test cases.

The BNG Blaster is completely built from scratch targeted for max scaling with a small resource footprint, simple to use and easy to integrate in our test automation infrastructure. It is able to simulate more than hundred thousand PPPoE and IPoE subscribers including IPTV, L2TPv2, QoS, forwarding verification and convergence testing capabilities.

• High Scaling: > 100K sessions
• Low CPU and Memory Footprint: < 100MB memory for 16K sessions
• Portable: runs on every modern linux, virtual machines and containers
• User Space: all protocols are implemented in user-space from scratch and optimized for high performance
• IPTV: IGMP version 1, 2 and 3 with automated channel zapping test
• QoS: define and analyze traffic streams
  Automation: the BNG Blaster Controller provides an automation friendly REST API and robot keywords
  ...

At RtBrick we already migrated all our testing to BNG Blaster which saved us more money on commercial network testers than spent on development for this software. We also received great feedback from GitHub users around the world using it. Therefore we will continue to sponsor this project with the target to form a large community of users and contributors.

In this presentation we will explore how Quantum Key Distribution (QKD) works and how it can be leveraged in existing security mechanisms. Before diving into QKD we will first look at some of the quantum terminology and principles. We will also explore what a Quantum Network and/or Quantum Internet is.

Tired of manually updating the ripe-db? What if the ripe-db is updated when you document your prefixes in netbox? Let's automate this on creation, update and deletion of objects. Dealing with corner cases such as overlapping prefixes is also covered. Backups included.

I identified all the IPv4 allocations by RIPE NCC to Germany, and did a programmatic survey of the state of reverse DNS for this data. I present my findings and talk very briefly about the benefits to network operators of establishing PTR records for external-facing nodes.

Booking.com runs large parts of its infrastructure on on-premise L3 Clos networks of non-trivial sizes. This talk presents our in-house end-to-end monitoring system that checks and reports on the health of the network, and helps us in checking our SLOs, and an ad-hoc tool used to debug ECMP issues in these networks.

Fast DNS resolvers are a crucial component in delivering a snappy internet experience for end-customers. If the resolvers are down or under heavy load, even the best internet connection is useless for most customers.
In this talk I'll cover how to design and set up a scalable, highly available DNS resolver infrastructure for internet service providers.
- Anycast IPs using BGP for horizontal scaling and high-availability
- Health checks
- Scaling nodes vertically to optimize cache ratios
- Automation is key

FTTH deployments and PONs (Passive Optical Networks) are coming up more often in discussions amongst network operators, and many times PON is used synonymous with inferior technology: only dark fiber is considered the one true solution.

As a network engineer with a strong Service Provider background, I can relate to this viewpoint, and while it is technically correct, the real world imposes financial, logistical and administrative restrictions on companies building and operating FTTH networks, thus making PON more beneficial for them.

This talk will give an overview of PON and PON designs (active and passive) and compares it with AONs (Active Optical Networks). Then it will show advantages and disadvantages of both technologies and give examples where one might choose one over the other technology.

End of Day 1

We'll run through a few Grafana developments you may have missed.

BGP local-preference is one of the most powerful attributes, but naively used, it inevitably gets in the way, and creates avoidable troubles for other networks.

This talk shows a few commonly used variants and the unwanted consequences.

There are several approved best practices and well-established procedures in aviation to reduce risk and guarantee safety. If you are running an operations team, there are some takeaways to improve your daily business.

This presentation describes recent enhancements to PING and TRACEROUTE as per RFC 5837 and RFC 8335. It also describes recent implementations.

In recent years so called "research networks", owned and operated by individuals, are on the rise. These networks are using public ASNs, IPv4 and IPv6 prefixes and are participating in the DFZ.

Whilst hands-on learning in general is great, we need to put this in context of the general stability of the internet at large, especially in current times, where routing security is not a default.

This talk offers the perspective of one individual network operator and aims to start a discussion, it is in no way to be interpreted as "the truth".

Find out how IXPs in Germany are developing

This presentation will demonstrate on how to leverage open standards like OpenConfig to provision BGP sessions in a vendor-agnostic approach. In addition the speaker will use the well-known PeeringDB API to obtain the peers information and present it in an OpenConfig-compatible format.

What Peering Manager(https://peering-manager.net/) can do for you and how to get started.

In this presentation, I will discuss how an SDN Controller that receives streaming telemetry from the network can automatically tune the paths of SR-TE or RSVP-TE LSPs in order to avoid congestion occurring. Contrary to popular belief, it is not necessary to fully mesh all of the edge nodes with TE LSPs to achieve this.

That's it, see you next time!