Today's session
Courses
- NVIDIA DOCA: zero to FAEBlueField DPUs, DOCA libraries and services, Dell PowerEdge qualification.0%0/43
- RoCE and RDMA for AI fabricsVerbs, RoCEv2 encapsulation, PFC/ECN/DCQCN, perftest and NCCL, and the counters that end the argument.0%0/27
- Spectrum-X Ethernet AI fabrics with Cumulus LinuxBuild, tune and troubleshoot an AI Ethernet fabric on NVIDIA Spectrum switches — and defend it in front of a Dell customer.0%0/29
- InfiniBand with QuantumOpenSM, UFM, partitions, adaptive routing and SHARP on NVIDIA Quantum fabrics.0%0/24
- Segment Routing over IPv6The SRH, uSID and the narrow slice of SRv6 that NVIDIA and Dell actually ship.0%0/12
- Kubernetes networking for AI clustersMultus, SR-IOV, RDMA in pods, the NVIDIA Network Operator, and the fabric under a Dell AI Factory.0%0/20
- Reference architectures and positioning: Dell AI Factory with NVIDIAWhich RA governs the deal, how to size the fabric from GPUs to BOM, and how to win Ethernet vs InfiniBand on published facts.0%0/20
Learn hard things properly. One course at a time.
A personal learning platform built on what the research actually supports: retrieval practice, spaced review, interleaving, worked examples that fade, and mastery gates. Every fact cites its source and the date it was checked.
NVIDIA DOCA: zero to FAE
BlueField DPUs, DOCA libraries and services, OVS-DOCA and HBN, DPF, Spectrum-X, Dell PowerEdge qualification, and customer scenarios. Interactive diagrams, two-variant labs (no hardware / BlueField-3 in the Dell lab), checkpoints.
Open the course mapRoCE and RDMA for AI fabrics
RDMA from the verbs object model up: how RoCEv2 puts InfiniBand transport on UDP/4791, how the ToS→DSCP→priority→traffic-class chain decides whether your traffic lands in the lossless queue, how PFC, ECN/DCQCN and ZTR-RTT congestion control actually close the loop, and how to prove any of it with perftest, nccl-tests and sysfs counters. Ends with Dell OEM field scenarios and an NCP-AIN coverage map. Interactive diagrams, two-variant labs (Soft-RoCE in containerlab / NVIDIA Air, or BlueField-3 and ConnectX in the Dell lab), checkpoints.
Open the course mapSpectrum-X Ethernet AI fabrics with Cumulus Linux
A hands-on course for a senior lab engineer moving into an NVIDIA Networking FAE seat on the Dell OEM account. It builds from the Spectrum ASIC portfolio and Cumulus Linux 5.18/NVUE fundamentals through RoCE QoS (PFC, ECN, buffer pools, packet trimming), the three Spectrum-X control loops (adaptive routing, telemetry-based congestion control, SuperNIC reordering and plane load balancing), telemetry and NetQ/WJH operations, and ends in FAE scenarios plus NCP-AIN blueprint prep. Every lesson ships two labs: one runnable with NVIDIA DSX Air, containerlab and Dell Enterprise SONiC 4.5.1, and one for the Dell lab's BlueField-3/ConnectX hosts, with customer-switch steps marked optional. Baselines: Cumulus Linux 5.18.1, Spectrum-X RA v2.3.1, NetQ 5.1.0, DOCA-Host 3.5.0-082.
Open the course mapInfiniBand with Quantum
InfiniBand from the subnet manager up: LIDs, GIDs and GUIDs; OpenSM, the switch-embedded SM and UFM; PKey partitions and SL/VL QoS; routing engines, adaptive routing, SHIELD and SHARP; rail-optimized fat trees on Quantum-2 (QM9700) and Quantum-X800 (Q3400); the infiniband-diags / ibdiagnet / mlxlink toolbox; and the Dell XE9680 side of an eight-rail pod. Every lesson ships two lab variants (no IB hardware, or the Dell-lab ConnectX / BlueField-3) and the course ends with FAE scenarios plus the InfiniBand and Troubleshooting domains of NCP-AIN.
Open the course mapSegment Routing over IPv6
SRv6 from RFC 8402/8754/8986 up through RFC 9800 compression, built and broken in containerlab with Linux seg6local and FRR, then mapped onto what really ships: Cumulus Linux 5.14+ uN/uA on Spectrum-4, DOCA Flow SRH push on BlueField-3/ConnectX, community SONiC HLDs, and a Dell Enterprise SONiC / OS10 feature matrix that publishes no SRv6 at all. Twelve lessons, three modules, every lab in a no-hardware (containerlab / NVIDIA Air) and a Dell-lab BlueField-3 variant, ending in the FAE conversation a Dell OEM account will actually have.
Open the course mapKubernetes networking for AI clusters
How an AI cluster actually wires GPUs together: the primary CNI (Calico/Cilium) that owns eth0 and Services, and the separate fast path — Multus, SR-IOV VFs, host-device, IPoIB, RDMA CNI — that carries NCCL traffic and never touches the primary CNI. Builds from the Kubernetes network model up through the NVIDIA Network Operator (v26.7.0), NIC Configuration Operator firmware templates, GPUDirect RDMA and NUMA alignment, to Spectrum-X rails, DPF, schedulers, triage and Dell AI Factory scenarios. Interactive diagrams, two-variant labs (containerlab/kind with no NVIDIA hardware, or the Dell-lab BlueField-3 and ConnectX hosts), checkpoints.
Open the course mapReference architectures and positioning: Dell AI Factory with NVIDIA
The design layer above the NIC. Read the published NVIDIA reference architectures — DGX SuperPOD, Enterprise RA (HGX AI Factory, NVL72 AI Factory), and the per-platform validated stacks — and know which one governs a given Dell deal. Decode node nomenclature (2-8-9-400), separate NVLink scale-up from the RDMA fabric, and size a cluster end to end: GPUs to SuperNICs to planes to leaf/spine ports to optics to BOM, with every ratio cited to the page it came from. Then map the RA onto what Dell actually sells (PowerEdge XE9680/XE9780/XE9785, PowerSwitch SN-series and Q-series, PowerScale F710), learn the three real support paths, and run the Ethernet-vs-InfiniBand positioning conversation on published claims only. Ends with timed FAE scenarios and an NCP-AIN gap plan. Interactive sizing tools, two-variant labs (containerlab/NVIDIA Air/Cumulus VX with no NVIDIA switch, or the Dell-lab BlueField-3 and ConnectX hosts).
Open the course mapToday's session
Cards that expire, the episode open on your desk and a drill on cases you have closed — in the order they should happen.
Review queue
Spaced repetition (FSRS). Cards come from lessons and from your quiz misses, across every course.
Dashboard
Mastery per course and module, calibration curve, weak spots, time on task.
Drill
Mixed retrieval interleaved across every course you have closed, or an interview drill weighted to the analysis and positioning questions an FAE actually gets asked.
Two or three questions before every lesson. Trying first, even wrong, makes the content stick (pretesting effect).
Short-answer before multiple choice, confidence rating before every reveal, misses become flashcards.
A Feynman box at the end of each lesson. Compare with a model answer, self-rate, build your own glossary.
Each module ends in a checkpoint. 80% unlocks the next one; failed items come back as targeted review.