In May 2016 the Packet Pushers hosted the Future of Networking Summit at Interop Las Vegas. On the second day of the summit, Ethan Banks sat down with network architect Russ White for an interview in front of a live audience about the current state of the network, emerging trends such as SDN and open networking, and why we need more protocols instead of fewer.
Ethan and Russ also took audience questions, which ranged from the relevance of industry certifications, to key skills for network engineers, to why thinking like a coder may be more important than learning to code. See the show notes for the complete list of questions.
Russ is a long-time industry pro who’s worked both at vendors and at customer sites and is well known as a networking subject matter expert. He’s authored or co-authored several books, and you’ll see his name at the top of many IETF RFCs. He’s also played a huge role in the design of several Cisco certification programs. He’s currently a network architect at LinkedIn.
The GNS3 Academy is offering the “Practical SDN and OpenFlow Fundamentals” course from David Bombal, and you can get 50% off. If you’re looking for a practical, hands-on, real-world approach to learning and implementing SDN and Openflow, look no further. The first 50 listeners to register receive the course for 50% off. Go to tinyurl.com/gns3sdn OR go to academy.gns3.com and use coupon: PACKETPUSHERS to get your discount.
Section 1 – How Are The Fundamentals Changing?
- Are we stuck with Ethernet forever?
- IPv6 seems like we boiled the ocean. Implementation is hard. Vendor support is unpredictable. Dual-stack is the easy part, really. What year will be the year of IPv6?
- Is HTTP the new TCP?
- In the unikernel movement, we strip out all extraneous anything – drivers, features, etc. – and have the barest possible environment that allows for an environment to interact with the outside world and run an application. Can networking learn from this? Are we too focused on backwards compatibility?
Section 2 – SDN
- We’ve been blathering on about SDN for 5 years now. Maybe more. How do you define SDN at this point, and what value do you think it will ultimately bring to the industry?
- Customers seem to want homogenous products. Products they can treat the same no matter what it says on the label. Can we ever make networking a predictable consumable?
- Does hyperconvergence absorb networking functionality in the long run?
Section 3 – Open Networking
- When is merchant silicon not good enough?
- Are there too many open networking projects for them all to survive? Can you suggest how we might pick out the winners?
- Standards bodies and open source seem at odds lately. Is that true, and will that remain true five years from now?
Section 4 – Audience Questions
- What about certifications? Are they viable for enterprise folks, for full-stack engineers, etc.?
- What are we doing to make applications work better with the network?
- How do you keep things simple when the requirements become complex?
- How do we deal with the firehose of complex things that keeps coming at us as network engineers?
- What key skills are there for the network engineer to pick up over the next 5 to 10 years?
- How do you learn to communicate with business people in their language?
- How can learning to think like a coder reduce complexity in your network?
- We have a lot of disruptive technologies, so where does that fit in with RFC 1925, rule 11 and the concept that nothing new is invented?
- You’ve said that BGP has become the “stuff protocol” because we’re stuffing too much stuff into it. Are you advocating for more protocols instead?
- Was there something unique about Ethernet and TCP that made them the winners?
- Will IEEE 1588 — synchronized Ethernet — ever go mainstream?
- Do you see firewalls in the future looking like they look today?