* Vativ Technologies doesn’t follow IEEE
A couple of weeks ago I wrote about a copper 10 Gigabit Ethernet transceiver from Vativ Technologies. That newsletter prompted a concerned response from the chair of the IEEE task force developing a standard around copper 10 Gigabit Ethernet.
Vativ Technologies’ approach should not be called “pre-standard” because “their chip is not even in the same ballpark as the 10GBase-T PHY being developed” by the IEEE 802.3an Task Force, wrote Chair Brad Booth, who is also Ethernet architect at Intel.
I wrote back asking whether it could even be called 10 Gigabit Ethernet, then. Booth answered, “It’s not 802.3 10 Gigabit Ethernet. It does transport Ethernet frames at 10G bit/sec, so one could call it 10 Gigabit Ethernet in the broadest sense of the term.”
I passed these concerns on to Vativ. In the end I got a perspective from Steve Kotowski, director of marketing for the company, who had some harsh words for the standards body on the issue of copper-based 10 Gigabit Ethernet. In the interest of fairness, I went back to Booth for a response. I’d like to present both sides in this newsletter and the next.
Kotowski wrote:
“The [transceiver] interfaces seamlessly into 802.3ae 10 Gigabit Ethernet through its standards-compliant XAUI interface. The transceiver operates autonomously from the MAC. Thus, the system will operate identically with our PHY as it would with a 10GBase-CX4 PHY, for example. Our PHY is proprietary, but it is completely interoperable with 10GbE.”
Booth responded:
“IEEE 802.3ae specifies port types, and compliance and interoperability testing is done at the connector. They might be running 10 Gigabit Ethernet in their device, but if you send a system based on their solution to the University of New Hampshire Interoperability Lab then it will not comply with any set of IEEE 802.3ae or 802.3ak tests. Nor will it comply with the IEEE 802.3an standard that is in development.”
More on this debate next time.
