This article will analyze two ConnectX-8 network card models with the same hardware and software baseline but with different port mapping and speed options. Provides a comprehensive overview of their technical variations and application scenarios to help network architecture teams make decisions that more closely fit their real-world networking needs.

Discover how 224G SerDes scales networks to 1.6T, breaking down signal integrity challenges and real-world 1.6T transceiver implementations.

Explore the evolution to 1.6T InfiniBand XDR: key tech drivers, hardware architecture, AI cluster deployment value, and practical interconnect selection.

As enterprises accelerate the adoption of AI, AI training and AI inference are increasingly appearing in the same infrastructure strategy. However, treating them as similar workloads can easily lead to pitfalls in network planning. In reality, the difference between training and inference goes far beyond the application level

The following content will analyze Physical AI from several dimensions: its technological evolution, underlying operating mechanisms, typical application scenarios, and NVIDIA's full-stack strategy.

The focus of supercomputing is gradually shifting from simply increasing computing power to the deep integration of various computing paradigms. As a key driver in the industry, NVIDIA has unveiled its next-generation scientific computing platform built around Simulation, AI (Artificial Intelligence), and Quantum Computing.

OSFP IHS (Integrated Heat Sink) is an OSFP packaging solution that integrates the heat dissipation structure directly into the module body. And OSFP-RHS (Riding Heat Sink) is an OSFP packaging solution designed for high-power optical modules and advanced thermal architectures, often referred to as Flat Top OSFP.

This article will introduce what XPO (eXtra-dense Pluggable Optics) is, how this module works, and its main differences and applications compared to CPO.

As AI and machine learning models keep growing, the need for high-speed connections in data centers is going up. The network bandwidth also needs to be increased. This makes it easier to use faster connection solutions, like 1.6T.

In the design of AI computing clusters, Scale-Up and Scale-Out have different goals from the outset, and their evaluation criteria are not aligned along the same dimension. This article will show what Scale-Up and Scale-Out is, and how to choose the optical module.