From the course: NVIDIA Certified Associate AI Infrastructure and Operations (NCA-AIIO) Cert Prep
Unlock this course with a free trial
Join today to access over 25,300 courses taught by industry experts.
ConnectX - NVIDIA Tutorial
From the course: NVIDIA Certified Associate AI Infrastructure and Operations (NCA-AIIO) Cert Prep
ConnectX
For all these AI supercomputers you are building, you would need networking, and that is where NVIDIA has solution for providing high-speed low-latency networking through InfiniBand and ConvertEthernets. This is ConnectX series. This came to NVIDIA's portfolio by acquisition of Metanox Technologies. Metanox was a well-known name in terms of InfiniBand software solutions and hardware solution. That was acquired by NVIDIA in 2022, I guess, and that's how NVIDIA was able to provide with all these InfiniBand SCAs and NICs. SCAs stand for Host Channel Adopter. It is high-performance interconnect adopter for InfiniBand, designed for ultra-low latency networking and high throughput, and it allows you to scale across AI, HPC, and enterprise data center workloads. It support advanced networking feature for GPU-accelerated system. In short, it is called RDMA. We will discuss about RDMA in the coming section. Primarily, it is used for AI training, HPC workload, data center where require fast…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
(Locked)
NVIDIA: Powering AI GPU innovation2m 37s
-
(Locked)
NVIDIA technology stack3m 12s
-
(Locked)
Layer 1: Physical layer3m 53s
-
(Locked)
GPU on a graphics card1m 57s
-
(Locked)
DGX platform2m 56s
-
(Locked)
DGX SuperPOD1m 57s
-
(Locked)
ConnectX1m 49s
-
(Locked)
BlueField DPUs2m 32s
-
(Locked)
NVIDIA reference architectures1m 38s
-
(Locked)
Understanding GPU cores5m
-
(Locked)
Comparing GPU cores4m 18s
-
(Locked)
NVIDIA DGX platform: Timeline4m 47s
-
(Locked)
DGX platform: Deployment options3m 38s
-
(Locked)
DGX A100 vs. H1004m 6s
-
(Locked)
Layer 2: Data movement and I/O acceleration59s
-
(Locked)
NVLink8m 5s
-
(Locked)
InfiniBand2m 5s
-
(Locked)
InfiniBand vs. Ethernet1m 43s
-
(Locked)
DMA and RDMA6m 30s
-
(Locked)
GPUDirect RDMA2m 44s
-
(Locked)
GPUDirect storage1m 45s
-
(Locked)
Quick comparison1m 56s
-
(Locked)
Layer 3: OS, driver, and virtualization2m 17s
-
(Locked)
GPU drivers4m 38s
-
(Locked)
GPU virtualization5m 8s
-
(Locked)
vGPU vs. MIG, part 17m 48s
-
(Locked)
vGPU vs. MIG, part 210m 59s
-
(Locked)
Layer 4: Core libraries6m 44s
-
(Locked)
Compute unified device architecture (CUDA)3m 12s
-
(Locked)
Installing CUDA2m 11s
-
(Locked)
NVIDIA collective communications library (NCCL)3m 41s
-
(Locked)
NVLink, NVSwitch, PCIe, RDMA vs. NCCL3m 44s
-
(Locked)
Layer 5: Monitoring and management2m 23s
-
(Locked)
NVIDIA-SMI4m 24s
-
(Locked)
Data Center GPU Manager (DCGM)7m 27s
-
(Locked)
Base Command Manager5m 33s
-
(Locked)
Which one to use?2m 3s
-
(Locked)
Layer 6: Applications and vertical solutions3m 48s
-
(Locked)
Summary2m 26s
-
(Locked)
NVIDIA AI Enterprise3m 2s
-
(Locked)
NVIDIA AI Factory2m 24s
-
(Locked)
-