Tag: nvidia
-
NVLink and NVSwitch: How NVIDIA Builds the Scale-Up Fabric (NVIDIA AI Series, Part 7)
Fifth-generation NVLink delivers 1.8 TB/s per GPU, and NVSwitch builds a non-blocking 130 TB/s all-to-all fabric across 72 GPUs in the GB200 NVL72. Here is how the domain forms, why it determines your tensor and expert parallelism strategy, and where the boundary falls.
-
GPU Partitioning on NVIDIA Data-Center GPUs: MIG vs vGPU vs Time-Slicing vs Passthrough (NVIDIA AI Series, Part 6)
Four ways to partition an NVIDIA H100, H200, or B200 GPU: MIG, vGPU, CUDA time-slicing, and full passthrough. This post covers the isolation guarantees, profile geometry, Kubernetes GPU Operator configuration, and a sizing worked example to help you pick the right mode for your cluster.
-
NVIDIA AI Factory Systems: DGX, HGX, MGX and NVL72 (NVIDIA AI Series, Part 5)
DGX, HGX and MGX are not performance tiers, they are three ways to integrate the same NVIDIA GPUs. Here is how they differ and where the GB200 and GB300 NVL72 rack actually earns its 120 kW.
-
GPU Memory and Precision: HBM3e, HBM4 and What Actually Fits (NVIDIA AI Series, Part 4)
A 70B model in FP16 needs 140 GB of weights before a single token of context. Here is the GPU memory and precision math that decides what fits, why HBM (not FLOPS) is the real ceiling, and where FP8 and NVFP4 buy you headroom.
-
NVIDIA Data-Center GPU Lineup: Hopper vs Blackwell vs Rubin (NVIDIA AI Series, Part 3)
The NVIDIA data-center GPU lineup from Hopper to Blackwell to Rubin, compared for training and inference: memory, bandwidth, FP4 and rack-scale NVL72, with a clear way to choose.
-
NVIDIA AI Enterprise: What the Subscription Includes and What It Costs (NVIDIA AI Series, Part 2)
NVIDIA AI Enterprise is the supported, secured wrapper around the open-source NVIDIA stack, licensed per GPU. Part 2 covers what is in the box (NIM, NeMo, Run:ai, the operators), how it is licensed (subscription, consumption, perpetual), what it costs, and when it is worth it.
-
What the NVIDIA AI Stack Actually Is, End to End (NVIDIA AI Series, Part 1)
NVIDIA AI is not one product, it is a stack roughly nine layers deep from silicon to agents. Part 1 maps the whole thing: GPUs, CUDA, the operators, TensorRT-LLM, Triton, Dynamo, NIM, NeMo, Nemotron, Blueprints and the AI Enterprise wrapper that supports it all.
-
Running GPU and AI Workloads on VKS (VKS Series, Part 14)
GPUs are where VKS stops being interchangeable with generic Kubernetes. Here is the vGPU VM class, the GPU Operator, and how VKS becomes the substrate for VMware Private AI.
-
GPU Partitioning for VMware Private AI: Choosing Between vGPU, MIG and Passthrough (Private AI Series, Part 6)
Time-sliced vGPU, MIG-backed vGPU, GPU passthrough and the new ESXi 9 Update 1 hybrid mode each fit different Private AI workloads. Here is how to design the split, with a capability matrix and a reference topology.
-
Choosing the Right GPU for VMware Private AI: L40S vs H100 vs H200 vs Blackwell (Private AI Series, Part 5)
A field-tested guide to picking the GPU for VMware Private AI Foundation: how L40S, H100, H200, RTX PRO 6000 Blackwell and A100 compare, why form factor beats the model name, and a clear verdict on which to choose for RAG, inference or training.
-
5 GPU & vGPU Mistakes That Break VMware Private AI Foundation (and How to Fix Them)
Most failed VMware Private AI Foundation deployments break on host-side GPU configuration, not the model. Here are five vGPU mistakes in VCF 9.1 and the exact commands to confirm and fix each one.
-
What is NVIDIA NeMo — and Why It Matters for Agentic AI
When people talk about AI systems, they often focus on models or APIs. But once you move beyond simple use cases, a bigger challenge appears: How do you control, guide, and manage AI behavior in real-world systems? This is where NVIDIA NeMo becomes critical. If NIM is the layer that runs AI models, then NeMo…
Architect’s Toolkit
PJ’s Tools
VMware Cloud Foundation
- VCF Documentation
- VCF 9 Planning & Preparation Workbook
- VCF Bill of Materials (BoM)
- VMware Compatibility Guide
- VMware Interoperability Matrix
- VMware Configuration Maximums
- VMware Ports & Protocols
- VMware Hands-on Labs
- RVTools Download
Nutanix
AI & Cloud-Native Platform
- NVIDIA Build (Model Catalog)
- NVIDIA AI Enterprise Reference Architecture
- NVIDIA NIM Performance Benchmarking
- NVIDIA NGC Catalog
- NeMo Microservices Helm Chart
- Helm Charts Repository
- Hugging Face Models
Architecture & Design
About the Author
Dr Pranay Jha
Dr. Pranay Jha is a Cloud and AI Consultant with 18+ years of experience in hybrid cloud, virtualization, and enterprise infrastructure transformation. He specializes in VMware technologies, multi-cloud strategy, and Generative AI solutions. He holds a PhD in Computer Applications with research focused on Cloud and AI, has published multiple research papers, and has been a VMware vExpert since 2016 and a VMUG Community Leader.
You May Have Missed
DrJha