- Podcast: ZigBits Network Design Podcast
- Episode: 67 – Top 5 Network Design Principles with Daren Fulwell
This episode of the ZigBits Network Design Podcast around network design fundamentals was the inspiration for me to start this blog series around podcast reviews. Either with work or study, I’m always finding myself thinking about design. I think it’s very important to not just understand the “what’s and how’s”, but also the “why’s” when it comes to any technology or design. Like I continuously say, it all starts with business requirements. I have yet to go through any network design focused certification or in depth study, so I thought it would be great to have some high level pillars in mind when thinking about network design that can be transferrable to different situations and scenarios. This episode gives us exactly that. Zig and Daren break down network design into five principles. Here is the list, with my spin/interpretation/thoughts of each. Zig has these listed out, with full explanations in the show notes.
- Availability
- I love that this is the first principle listed and it makes complete sense. How much value can a network provide if it is only available and functioning a small portion of the time? It is highlighted in the podcast that availability is part of security minded CIA triangle of confidentiality, integrity, and availability.
- Scalability
- This tends to be a tough one for me. Is there a “magic” percentage number that we should select for every design in terms of growth? For example, when it comes to switches and routers, should I always account for 25% port count growth? Or 30%, or even 50%?! Let’s face it, planning for a fair amount of unknowns is difficult, but we need to do our best with the information given. My takeaway from this episode is that the keys are elasticity and the ability for the network to be flexible as business needs change.
- Security
- Security consideration really should not/cannot be left out in the design and implementation processes. Like mentioned earlier, it is directly related to availability.
- Supportability
- This is incredibly important, but I could see how it might sometimes be an afterthought. There could be a completely solid design on the table, but if it takes someone with 20 years of experience and 15 certifications to decipher it and troubleshoot when issues arise, is it really a solid design after all? Shifting gears a little bit, one thing that was brought up in the episode was automation and templates. If configuration standardization is a consideration in the design phase, automation can more easily be leveraged to support the design not only at implementation, but over time as well.
- Simplicity
- To me, this goes hand-in-hand with the supportability principle. Care should be taken when technologies and features are being considered for implementation. Maybe a cost/benefit analysis needs to be done whenever something being considered for deployment adds a certain level of complexity. At the end of the day, whatever is being implemented has to be supported by someone. I think I may have mixed thoughts a bit in these last two principles in relation to the episode, but I think these two are tightly coupled.
Final Thoughts
Again, this podcast episode was the inspiration for me to start this series of posts. I tend to have the desire to be “structured”, and I’m a creature of habit. To be able to look at network designs in the future with these five principles fresh in mind is incredibly helpful. Zig and Daren clearly know what they are talking about around this topic and have the CCDE certificates to prove it! This was a great listen and I strongly recommend it to any network engineer/admin/architect.
Netication 