CCIE #69931 · AI fabrics, data center, low latency, automation
Network designs, proven in the lab.
Writing and labs on networks that can't afford guesswork: GPU fabrics for AI training and inference, data center, low-latency trading and automation. Every design comes with the lab output and packet captures behind it, so you can check the work.
Two ways in
For engineers
Learn it, then prove it in a lab
Explainers that start with why, labs you can rebuild with the configs, output and packet captures to check your results against, and network tools for subnets, multicast MACs and MTU.
For certification candidates
Study the blueprint, domain by domain
Two study series that follow the published blueprints for NVIDIA's NCP-AIN and the CCDE AI Infrastructure track: the reasoning behind each topic, the numbers worth knowing and the commands to practise, one part at a time and in order.
Explore by topic
5 articles · 5 study parts
AI infrastructure
GPU fabrics for training and inference: RoCE and InfiniBand, congestion, sizing and reference designs.
2 articles
Data center & enterprise networking
BGP, EVPN and fabric design for the networks everything else runs on, built and broken in labs.
- Designing an EVPN-VXLAN fabric across two vendors, step by step
- BGP unnumbered, explained without the hand-waving
- Decision: ASN plans for leaf-spine and rail-optimized AI fabrics Coming 2 Dec
2 articles
Low latency & trading
Market data, multicast and the networks where microseconds and missing sequence numbers cost money.
First articles coming soon
Automation
Configuration as code, verification and the tooling that keeps a fabric consistent.
Latest articles
NCP-AIN Spectrum domain: RoCE on Spectrum-X, QoS, ECN and PFC end to end
Part 4 of the NCP-AI Networking study series: one RoCE packet followed from the sending SuperNIC through every Spectrum switch to the receiver and back, with the DSCP, priority and traffic-class contract every hop must share, the NVUE and host commands behind it, PFC headroom and the watchdog, and the counter patterns that tell you which hop disagrees.
CPU, GPU and TPU: what each one does, and how they work together in an AI cluster
Why AI runs on GPUs and TPUs rather than CPUs, what is different inside each chip, their pros and cons, and how CPUs, accelerators, memory and the network cooperate in every training step.
A trading day in multicast: SSM, ASM, BIDIR and every way the feed breaks, proven on IOS-XR and NX-OS
Follow one trading day at a fictional firm, from the feeds coming up before the open to the post-mortem after the close. Each incident explains a piece of trading multicast (A/B feeds and sequence numbers, SSM, ASM and the switch to the source tree, SSM mapping, pre-joined trees, BIDIR with a phantom RP, Anycast RP, the extranet boundary, IGMP snooping queriers, microbursts) and then proves it in an IOS-XR and NX-OS lab with sequence-numbered traffic, CLI output and packet captures.
Coming next
The next articles on the schedule. The certification series publish part by part, in order.