DriveNets Blog

After spending time at RAISE Summit 2026 in Paris, one message was clear: the AI industry is moving from model...

The AI industry has grown significantly since the introduction of the attention mechanism in 2017, reaching the point where models...

We previously discussed an operational reality for NeoClouds and AI infrastructure builders. AI workloads are increasingly limited by the power,...

We’re living through the most active AI accelerator cycle in the AI era. NVIDIA Blackwell is here. Rubin’s next. AMD is...

Artificial intelligence is here to stay. The ever exponentially-growing investment in AI infrastructure is expected to continue, as more and...

If you have built an AI cluster, you already know it is not a simple task. It is not just...

AI workloads have moved beyond experimentation into deep production environments. Today, performance is no longer about peak FLOPS (floating-point operations...

AMD Instinct GPUs offer real hardware advantages for large AI clusters, including higher HBM3 memory capacity and competitive compute density....

First off, may the force be with you on this May the Fourth. The most oft-repeated line, “May the Force be...

For years, the AI industry operated on one rule: compute is king. More parameters, more GPUs, more raw processing power....

As AI and HPC applications diversify at breakneck speeds, the “one cluster, one job” model is fading. ...

AI has been dominating technology-related conversations for over two years now. This also applies to the service provider (SP) market,...

Last week, I celebrated 5 years in DriveNets (joined in January 2021, in the middle of the Covid-19 pandemic). Even...

Today it’s clear to everyone in the AI infrastructure industry that training massive models isn’t just about raw compute power,...

Today, the rapid expansion of AI infrastructure is introducing new challenges that stretch the limits of both traditional data centers...

In the high-stakes world of global finance, the “race to zero latency” has evolved. For years, the battleground for high-frequency...

Heading back from St. Louis late last week, I had some time to reflect on the very busy week of...

Carved your pumpkin and picked the perfect Halloween costume? Halloween is not just a good time for sharing scary stories,...

As AI clusters scale from thousands to hundreds of thousands of GPUs, the choice of network fabric, InfiniBand or Ethernet,...

It’s that time of year again – time to recap the annual OCP Global Summit that was recently held in...

The DriveNets AI Fabric networking fabric solution delivers the highest performance AI connectivity for any GPU, NIC or optics, based...

Have you noticed how “scale-up” and “scale-out” are becoming popular terms in recent AI infrastructure discussions? This is not a...

Building AI infrastructure goes far beyond simply connecting GPUs, it demands precise orchestration of compute, networking, storage, and system configuration...

What do you need to know about Andy Holland – DriveNets’ new AI sales leader? –> With excellent solutions and experienced...

Strong partnerships that substantially increase value provided to customers can fundamentally change a company’s – and even an industry’s –...

I just got back from rainy Hamburg after another great ISC conference. ISC 2025 was an event full of energy...

Parallelism is the engine behind modern AI scalability, but it’s only as effective as the networking infrastructure supporting it. The...

Today high-performance computing is no longer confined to hyperscalers. NeoClouds and Enterprises across finance, healthcare, telecom, and other industries are...