Will 10Gb Ethernet take off in 2015?
Intel promised $50 10Gb Ethernet ports back in 2012. But you rarely see them even on workstation class machines.
Adoption has been slowed by several issues:
Gigabit Ethernet works fine for most PCs, and it's PCs that drive volume. No volume, no learning curve, no investment.
Expensive wiring, connectors and chip sets have limited 10GigE to high-end applications, like big servers hosting multiple VMs.
Most processors couldn't drive a 10Gig channel because of architectural limitations.
Connection cost
The biggest stumbling block was the requirement for SFP+ connectors. These small form-factor pluggable connectors and their cabling were not compatible with the RJ45 connectors used on 1GbE networks.
Special Feature
Storage: Fear, Loss, and Innovation in 2014
The rise of big data and the demand for real-time information is putting more pressure than ever on enterprise storage. We look at the technologies that enterprises are using to keep up, from SSDs to storage virtualization to network and data center transformation.
Few SFP+ cables cost less than $100 and many are much more. That's way more than the $40 Thunderbolt cables people complain about.
But relief is on the way: 10GbE that runs over Cat 6 and 7 wiring using RJ45 connectors that we already use everywhere. Combined with lower cost 10GbE chipsets and improved CPU I/O, we have the ingredients for a 10GbE resurgence.
The Storage Bits take
10GbE networks will start out in datacenters, but workstation-class PCs should start adding the ports in the next 2 years. Higher speed Ethernet will be stiff competition for Fibre Channel SANS and older Infiniband networks.
Home adoption is still 5-10 years away. Home users prefer Wi-Fi for the convenience and most home connections can't even manage 100Mbs.
But the advent of 4k TV will push more home users to higher bandwidth. It will happen, but slowly.
Comments welcome, as always. I like 20Gb Thunderbolt a lot, but I'm not sure when I'll need a 10Gb network. What do you think?
