Speaking metaphorically, the blast area of a datacenter should only be so large, for mechanical as well as technical reasons, in order to mitigate the risk of a failure taking out too many machines. …
Jayshree Ullal, the chief executive officer of Arista Networks, has been pointing at the $10 billion revenue upper decks since the beginning of the AI boom, but is understandably hesitant about saying when the company would cross that threshold. …
InfiniBand was always supposed to be a mainstream fabric to be used across PCs, servers, storage, and networks, but the effort collapsed and the remains of the InfiniBand effort found a second life at the turn of the millennium as a high performance, low latency interconnect for supercomputers running simulations and models. …
After three relatively short years of explosive growth thanks to the GenAI boom, AI is driving half of systems revenues worldwide already. …
While it has always been true that flatter networks and faster networks are possible with every speed bump on the Ethernet roadmap, the scale of networks has kept growing fast enough that the switch ASIC makers and the switch makers have been able to make it up in volume and keep the switch business growing. …
The rapid changes Broadcom instituted after buying virtualization stalwart VMware for $61 billion in late 2023 continue to shape the virtualization and cloud spaces, with some enterprises facing significant higher pricing, new licensing plans, and bunding options looking for alternatives, vendors offering them alternatives, and companies rolling out plans to help with the migrations. …
The semiconductor manufacturing business is absolutely immense. To give the numbers some perspective, in 2024, chip makers generated revenues that were about three quarters of the size of the US defense budget and about two-thirds the size of the social services budget allocated by Congress. …
Compute engine makers can do all they want to bring the performance of their devices on par or even reasonably close to that of Nvidia’s various GPU accelerators, but until they have something akin to the NVLink and NVSwitch memory fabric that Nvidia uses to leverage the performance of many GPUs at bandwidths that dwarf PCI-Express switches and latencies that dwarf Ethernet interconnects, they can never catch up. …
Nvidia sells the lion’s share of the parallel compute underpinning AI training, and it has a very large – and probably dominant – share of AI inference. …
All presidents of these United States have the bully pulpit from which to lecture the American people and, for the past century, the rest of the world about how the global economy and culture should work. …
All Content Copyright The Next Platform