AIC-200 · GPU & AI Compute
Linux for GPU Systems
What the kernel is doing underneath your training job, and how to tune it. The layer almost nobody teaches.
Who this course is for
Platform and infrastructure engineers responsible for the Linux layer under GPU workloads — drivers, scheduling, memory, and the performance tuning that decides whether expensive GPUs are actually used.
Prerequisites
Course outline
Day 1 — Linux kernel internals I
- Kernel architecture and subsystems
- System call mechanism from user to kernel
- Kernel module API: init/exit, symbol export
- Process scheduling: CFS vruntime, red-black tree, load balancing
- RT policies SCHED_FIFO/SCHED_RR
Day 2 — Linux kernel internals II
- CPU isolation: isolcpus, nohz_full, rcu_nocbs
- Memory management: mmap, VMA, page faults, buddy allocator, slab, vmalloc
- Memory pressure: kswapd, OOM, cgroup v2
- Device model, bus_type and sysfs
- Kernel build and configuration; debugging with printk/dynamic debug
Day 3 — Linux performance tuning
- CPU affinity: taskset, sched_setaffinity; NUMA-aware launch with numactl
- Huge pages and TLB miss analysis
- I/O schedulers (none, mq-deadline, kyber, bfq) for NVMe
- Network stack tuning: sysctl, BBR, busy polling
- Profiling with perf, ftrace, eBPF/bpftrace/BCC
- IRQ affinity, tickless config, RCU offloading, latency histograms
Day 4 — Linux for GPU workloads
- NVIDIA driver stack: nvidia.ko, nvidia-uvm.ko, nvidia-modeset.ko
- Driver installation: .run, packages, DKMS; module signing
- DRM/KMS; PCI enumeration with lspci -vvv; /dev/nvidia*
- GPU passthrough: VFIO, IOMMU, QEMU; SR-IOV for NICs
- NVIDIA Container Toolkit and runtime hooks
- Driver troubleshooting: dmesg, nvidia-bug-report.sh
Hands-on labs
Labs follow the academy model — 35% principles, 20% guided investigation, 45% engineering studio. Every claim you make in a lab is backed by a trace, a counter or a measurement you captured yourself. How we teach
- Lab: build and load a kernel module; navigate the scheduler and memory code paths it touches
- Lab: isolate CPUs with isolcpus/nohz_full/rcu_nocbs and prove jitter reduction on a latency histogram
- Lab: profile a workload with perf and an eBPF/BCC tool; produce a flame graph and act on it
- Lab: install and validate the NVIDIA driver stack; pass a GPU through to a QEMU VM with VFIO
- Lab: configure the NVIDIA Container Toolkit and verify GPU access from inside a container
Capstone project
Take a misconfigured GPU server from symptom to tuned baseline: discover the driver, scheduling, memory and IRQ problems you are given, fix them with evidence, and deliver a reproducible tuning manifest with before/after latency and throughput distributions.
What you leave with
- A repeatable kernel tuning methodology, not a bag of sysctl tricks
- Working skills with perf, ftrace, eBPF/BCC and flame graphs
- NVIDIA driver lifecycle and VFIO passthrough experience
- A tuning manifest template applicable to your own fleet
How it runs
Every course follows the same model: 35% principles, 20% guided investigation, 45% engineering studio. You leave with working code, raw measurements and an evidence-based report — not a certificate of attendance. Read the methodology or see a full sample lesson.
Material is adapted to your kernel version, hardware and workload before a private delivery. For public cohorts, the environment is provided and configured.
Questions
Who is this course for?
Platform and infrastructure engineers responsible for the Linux layer under GPU workloads — drivers, scheduling, memory, and the performance tuning that decides whether expensive GPUs are actually used. It sits at practitioner level within the GPU & AI Compute track.
What do I need to know already?
Specific prerequisites for this course: Solid Linux administration; AIC-100/AIC-110 or equivalent knowledge; Ability to read C (kernel source reading is guided). We confirm levels before the cohort starts and adapt if a group is stronger or weaker than expected.
Can this run privately for my team?
Yes. Any course runs on-site at your offices anywhere, or live online for a distributed team, with labs adapted to your hardware and codebase.
What is the difference between in-person and online?
In person is 4 full days with hardware on your desk, capped at 14. Online is 8 half-day sessions across about two weeks so you can keep working, capped at 20, with remote lab access.
Do you invoice companies?
Yes. Purchase orders are accepted and invoicing is available in USD, EUR, GBP, SAR and CAD.
Upcoming dates
| Dates | Where | Seats | Early bird | Regular | |
|---|---|---|---|---|---|
| 15 Nov – 18 Nov 20264 full days | RiyadhIn person · KAFD Conference Centre | 11 of 14 | SAR 9,450until 16 Oct | ||
| 22 Nov – 25 Nov 20264 full days | Kuwait CityIn person · Al Hamra Tower | 6 of 14 | KWD 780until 23 Oct | ||
| 22 Nov – 25 Nov 20264 full days | MuscatIn person · Knowledge Oasis Muscat | 11 of 14 | OMR 970until 23 Oct | ||
| 29 Nov – 8 Dec 20268 half-days | Gulf bandLive online · 09:00–13:00 GMT+3 | 3 of 20 | US$1,800until 30 Oct | ||
| 30 Nov – 3 Dec 20264 full days | OttawaIn person · Kanata North Tech Park | 6 of 14 | CAD 3,430until 31 Oct | ||
| 7 Dec – 10 Dec 20264 full days | TorontoIn person · MaRS Discovery District | 11 of 14 | CAD 3,430until 7 Nov | ||
| 7 Dec – 10 Dec 20264 full days | LondonIn person · Shoreditch Works | 6 of 14 | GBP 1,960until 7 Nov | ||
| 7 Dec – 16 Dec 20268 half-days | Europe bandLive online · 09:00–13:00 CET | 8 of 20 | US$1,800until 7 Nov | ||
| 7 Dec – 16 Dec 20268 half-days | Americas bandLive online · 13:00–17:00 ET | 13 of 20 | US$1,800until 7 Nov | ||
| 14 Dec – 17 Dec 20264 full days | BerlinIn person · Factory Görlitzer Park | 11 of 14 | EUR 2,320until 14 Nov |
Dates shown for the next few months. If nothing fits, tell us where and when — cohorts are added on demand, and private delivery can be scheduled any week.
More in GPU & AI Compute
AIC-1003 days
Foundations for AI Compute
The architecture, operating system and networking groundwork every GPU systems engineer is assumed to have and often does not.
Practitioner-taught
SAR 6,750Next 25 Oct
AIC-1103 days
GPU Architecture, Memory & Interconnects
How the hardware constrains your workload: SIMT execution, the memory hierarchy and the fabric between GPUs.
Practitioner-taught
SAR 6,750Next 11 Oct
AIC-2104 days
RDMA & AI Cluster Networking
Build and debug the fabric distributed training runs on, from queue pairs up to a tuned NCCL all-reduce.
Practitioner-taught
SAR 10,500Next 1 Nov
AIC-2205 days
CUDA & HIP Programming
Write, profile and optimise GPU kernels on both vendors, including the CUDA-to-HIP porting path.
Practitioner-taught
SAR 13,120Next 11 Oct
AIC-2304 days
Distributed Training with PyTorch
From a single-GPU training loop to sharded multi-node training that survives a node failure.
Practitioner-taught
SAR 10,500Next 15 Nov
AIC-3004 days
Containers, Kubernetes & GPU Schedulers
Run a shared GPU cluster multiple teams can actually use: partitioning, scheduling, quotas and isolation.
Practitioner-taught
SAR 10,500Next 18 Oct
AIC-3103 days
Storage & Data Pipelines for AI
Stop starving your GPUs: parallel filesystems, GPUDirect Storage and pipelines built for sustained throughput.
Practitioner-taught
SAR 7,880Next 22 Nov
AIC-3203 days
MLOps & Inference Serving
Get models off a laptop and onto a GPU endpoint that scales, with the pipeline machinery around them.
Practitioner-taught
SAR 7,880Next 1 Nov
AIC-3302 days
AI Infrastructure Security & Observability
Harden a multi-tenant GPU platform and see what it is doing before users report a problem.
Practitioner-taught
SAR 6,000Next 18 Oct
AIC-4003 days
Large-Scale Training & Datacenter Architecture
The thousand-GPU conversation: 3D parallelism, reference architectures, TCO and where the hardware is heading.
Practitioner-taught
SAR 9,000Next 8 Nov