NCCL

Oct 20, 2025

Scaling Large MoE Models with Wide Expert Parallelism on NVL72 Rack Scale Systems

Modern AI workloads have moved well beyond single-GPU inference serving. Model parallelism, which efficiently splits computation across many GPUs, is now the...
10 MIN READ
Sep 09, 2025

How to Connect Distributed Data Centers Into Large AI Factories with Scale-Across Networking

AI scaling is incredibly complex, and new techniques in training and inference are continually demanding more out of the data center. While data center...
6 MIN READ
Jul 22, 2025

Understanding NCCL Tuning to Accelerate GPU-to-GPU Communication

The NVIDIA Collective Communications Library (NCCL) is essential for fast GPU-to-GPU communication in AI workloads, using various optimizations and tuning to...
14 MIN READ
Jul 18, 2025

Optimizing for Low-Latency Communication in Inference Workloads with JAX and XLA

Running inference with large language models (LLMs) in production requires meeting stringent latency constraints. A critical stage in the process is LLM decode,...
6 MIN READ
Jul 14, 2025

Enabling Fast Inference and Resilient Training with NCCL 2.27

As AI workloads scale, fast and reliable GPU communication becomes vital, not just for training, but increasingly for inference at scale. The NVIDIA Collective...
9 MIN READ
Jul 14, 2025

NCCL Deep Dive: Cross Data Center Communication and Network Topology Awareness

As the scale of AI training increases, a single data center (DC) is not sufficient to deliver the required computational power. Most recent approaches to...
9 MIN READ
Jun 18, 2025

Improved Performance and Monitoring Capabilities with NVIDIA Collective Communications Library 2.26

The NVIDIA Collective Communications Library (NCCL) implements multi-GPU and multinode communication primitives optimized for NVIDIA GPUs and networking. NCCL...
11 MIN READ
Typical data center interconnection schema for Clos fabric.
May 14, 2025

AI Fabric Resiliency and Why Network Convergence Matters

High-performance computing and deep learning workloads are extremely sensitive to latency. Packet loss forces retransmission or stalls in the communication...
7 MIN READ
Mar 13, 2025

Networking Reliability and Observability at Scale with NCCL 2.24

The NVIDIA Collective Communications Library (NCCL) implements multi-GPU and multinode (MGMN) communication primitives optimized for NVIDIA GPUs and networking....
14 MIN READ
Jan 31, 2025

New Scaling Algorithm and Initialization with NVIDIA Collective Communications Library 2.23

The NVIDIA Collective Communications Library (NCCL) implements multi-GPU and multinode communication primitives optimized for NVIDIA GPUs and networking. NCCL...
9 MIN READ
Picture of servers in a data center.
Oct 25, 2024

Advancing Performance with NVIDIA SHARP In-Network Computing

AI and scientific computing applications are great examples of distributed computing problems. The problems are too large and the computations too intensive to...
7 MIN READ
Decorative image of a cube of green cubes, surrounded by other cubes on a dark background.
Sep 16, 2024

Memory Efficiency, Faster Initialization, and Cost Estimation with NVIDIA Collective Communications Library 2.22

For the past few months, the NVIDIA Collective Communications Library (NCCL) developers have been working hard on a set of new library features and bug fixes....
8 MIN READ
Sep 06, 2024

Enhancing Application Portability and Compatibility across New Platforms Using NVIDIA Magnum IO NVSHMEM 3.0

NVSHMEM is a parallel programming interface that provides efficient and scalable communication for NVIDIA GPU clusters. Part of NVIDIA Magnum IO and based on...
7 MIN READ
Apr 26, 2024

Perception Model Training for Autonomous Vehicles with Tensor Parallelism

Due to the adoption of multicamera inputs and deep convolutional backbone networks, the GPU memory footprint for training autonomous driving perception models...
10 MIN READ
Mar 06, 2024

CUDA Toolkit 12.4 Enhances Support for NVIDIA Grace Hopper and Confidential Computing

The latest release of CUDA Toolkit, version 12.4, continues to push accelerated computing performance using the latest NVIDIA GPUs. This post explains the new...
9 MIN READ
Oct 12, 2023

Networking for Data Centers and the Era of AI

Traditional cloud data centers have served as the bedrock of computing infrastructure for over a decade, catering to a diverse range of users and applications....
6 MIN READ