25 GbE not on the horizon
April 9th, 2014
Some industry leaders have believed that the IEEE would be well served creating engineering and network cabling standards to support 25 Gbps Ethernet functionality for data centers and similar environments. According to a recent Lightwave article, discussions about 25 GbE reached a tipping point at a recent IEEE event, but they didn't end up going anywhere. After some time , the talks dissolved with parties deciding that there was not even enough support to make it worthwhile to call for a vote on the possibility of creating a 25 Gbps Ethernet working group.
Looking at the possibility of 25 GbE
The article explained that some industry stakeholders hoped that 25 GbE functionality could serve as a good stepping stone for organizations that need better performance than 10 Gbps solutions, but are not yet ready to make the jump to a fully featured 100 Gbps setup.
While a jump from 10 Gbps to 100 Gbps may seem staggering, the IEEE is already working toward 400 Gbps Ethernet functionality, so moving to focus on 25 GbE as well could have been too much.
According to the news source, some of the impetus to move to 25 Gbps Ethernet came because the performance specifications already exist in some form. The IEEE standard for 100 GbE solutions are built around connecting four 25 Gbps channels into a unified setup that generates performance capabilities reaching 100 Gbps speeds. As such, the thought was that it would be fairly natural to create a standard for 25 Gbps functionality because the process already exists.
Sources told Lightwave that while there were some high-profile supporters for the move to 25 Gbps Ethernet, there were some stakeholders in the IEEE that were worried that any move to that performance standard would detract key resources from 400 Gbps network standard development.
Considering standard development in a broad scale
Network performance standards do not always develop in a linear scale, moving from one performance benchmark to the next. The reality is that standardization is usually years ahead of the point where businesses will actually be prepared to invest in the new solutions at any large scale.
A few early adopters may be ready once a standard is released, but most are behind the curve because costs can be high when dealing with cutting-edge solutions. As such, it is not shocking when some industry participants want a standard that is for performance levels below the peak of the market. The thought is that these standards would focus more on making advanced technologies accessible. At this point, however, it seems that the IEEE is prioritizing future progress.