Skip to main content

Posts

Packetpushers - Big Rock or Best-in-Breed (heavy strategy)

I recently listened to a conversation on Packet Pushers that really got me thinking. In the past, I’ve seen a similar pattern in large transformations where organizations adopt the “Big Rock” approach—going all in on a single, massive solution to address their needs. Each time, it didn’t deliver the expected results. So why does this approach keep getting chosen? Is it driven by cost-saving measures, the desire for streamlined management, or is there something else at play? Of course, the best-of-breed approach has its own set of challenges—like managing multiple vendors or integrating different systems—but doesn’t it offer a safer path for better business continuity and outcomes? It allows for flexibility and adaptability, which are crucial in today’s fast-changing landscape. While it’s not without its headaches, it seems like a more reliable bet in the long run. Link to the talk . 

In love with Golang

I’ve always been the type of person who enjoys a good challenge—whether it’s figuring out a new skill or diving into something I have no idea about. Recently, my latest obsession has been cloud-native programming. I got involved in a project that’s all about acquiring cloud-native applications for telecoms, and it felt like the perfect excuse to push my limits. Now, let’s be clear—I don’t code every day. I don’t even get the chance to write code as often as I’d like. And when I do, it takes me a minute to remember how things work.  Here’s the thing—I know I’ll never be a “proper” programmer and I’m not trying to be. But that doesn’t stop me from pushing myself. My goal is simple: keep learning, keep getting better, and see how far I can take it. One thing I’ve discovered along the way is how much I love Golang. I’ve dabbled in Python, but Golang just clicks for me. It’s that sweet spot where things make sense and don’t feel as clunky. It even helped me have some solid technical con...

IPv6 Council London - 2022 (Meta)

I truly enjoyed the last IPv6 Council meeting in London. It's always interesting to hear about the other people experience with a technology, in this case.. IPv6.  There were quite a few people giving a presentation and how they IPv6 deployment is going, so that was interesting!   I personally have been pushing the IPv6 agenda and I consider myself as an IPv6 advocate, because I understand the business need for it, and most of the people I have deployed carrier-grade NAT (large scale), and I also remember when I took the last IPv4 range from RIPE.  I guess, I will finish this post with - if anyone is interested in IPv6, the IPv6 Council London is free, and it's quite interesting! Link: https://www.ipv6.org.uk/

DC with Arista Networks

I got this fantastic opportunity to deploy new clouds based on Arista Networks VXLAN fabric. The solution included Arista CloudVision (known as CVP), AVD and of course, the Arista switches. The first thing I was impressed with was the simplicity that Cloud Vision brings on the table for both deployment of new configurations and visibility.  We were waiting for the hardware to arrive, and all we got was the CVP installed and ready, so we did a bit of work e.g. getting the configlets ready and when the hardware arrived we plugged all the optics and connected the fibres. In that moment one of my colleagues who didn't have experience with Arista decided to go around and check for any obvious errors, so he did not find anything than just couple of links require fibre change, not a big deal. On the next day, we start exploring what can be observed via CVP, and we found it actually start reporting packet drops/ and errors via the telemetry that is supported via Terminattr telemetry agent ...
It's been almost 2 years since I joined Juniper's SD-WAN "Contrail Service Orchestration" project and I am a bit sad leaving some great people behind and it was a tough decision but I couldn't spent enough time with my family.  I will definitely miss my time with Juniper Networks. 

python

always wanted to do more coding and I was doing some of the basic challenges on hackerrank recently and I am so disappointed in myself; I haven't written anything for a while and now struggling.

JNPR

wow, it's been more than a year since I joined this new opportunity working on service orchestration project for SD-WAN and it's great, as I am doing a bit more hand-on work. I kind of missed the 'console' if I can use that expression.. :)