For whatever reason, it takes a lot longer to case the commercial Ethernet switching and routing markets, which is dominated by datacenter and campus use cases, than it does for the server and storage markets. …
As supercomputer centers have long known and as hyperscalers and cloud builders eventually learned, the larger the cluster, the greater the chance that one of the many components in the system will fail at any particular time. …
Amazon Web Services, the juggernaut of cloud computing, may be forging its own path with Arm-based CPUs and associated DPUs thanks to its 2015 acquisition of Annapurna Labs for $350 million. …
The hyperscalers and cloud builders can be split into two camps, but a third one might be emerging. …
If the datacenter had been taken over by InfiniBand, as was originally intended back in the late 1990s, then PCI-Express peripheral buses and certainly PCI-Express switching – and maybe even Ethernet switching itself – would not have been necessary at all. …
Cisco Systems may still be the biggest supplier of switches and routers in general, but it has long since been surpassed by Broadcom when it comes to suppling the silicon that does the switching itself and sometimes even a little bit of routing in the datacenter in particular. …
If you want to get a sense of what is happening in the high-end of the Ethernet switch and routing market, it is Arista Networks, formerly an upstart and now just one of the bigger vendors taking on the hegemony of Cisco Systems in networking in the datacenter and now on the campus and at the edge, is probably the best bellwether there is. …
It takes money to make money, and if you want to break into the switch ASIC business in the datacenter, even if you are a low-cost designer of such chips, you had better have some rich friends to help the business take off. …
Networks may not be the most expensive thing in the datacenter – they typically comprise about 10 percent to 15 percent of the cost of a distributed system, including cables, transceivers, switches, and routers – but they are without a doubt the most complex part of distributed systems. …
It is hard to remember sometimes way back when, in 2008, as Nvidia first took a stab at GPU compute in the datacenter with the original Tesla GPU accelerators and a very rudimentary CUDA programming environment for offloading parallel algorithms from CPUs to GPUs. …
All Content Copyright The Next Platform