According to at least a couple of product and component manufacturers in the networking field, two of the obstacles that initially make 10GBase-T a tough sell are on their way down.
One, shockingly (note the sarcasm in my keystrokes), is cost. The other is power consumption.
A 10GBase-T switch was introduced in May with a manufacturer’s ballpark price of “a little more than 2x” a 1000Base-T switch. Admittedly, manufacturer Arista Networks is positioning the switch as a low-cost option. So, don’t expect the same price point from the other 10GBase-T switch(es) on the market. As I write this in mid-June, I know of just one other such switch available for purchase. By the time you read it, there may be others.
The other obstacle is power consumption. Followers of 10-Gigabit Ethernet technology have been hearing for a long time that the transmission speed’s Base-T variety is a power glutton. A recent discussion, how- ever, with the chief technology officer at physical layer interface (PHY) provider Solarflare Communications gave me a peek into what is “in the oven,” as he put it, with original equipment manufacturers regarding power consumption.
“We’re sitting at a really interesting time,” says Solarflare’s George Zimmerman. To summarize extremely briefly, in 12 to 18 months, we should see 10GBase-T products coming to market that consume far less power than today’s devices require.
After conversations with Arista and Solarflare, I Tweeted on the idea that these two hurdles are being lowered, likely prompting 10GBase-T adoption. The responses that resulted were so ... Twitter. For example:
“Until the power consumption gets very close to fiber, 10GBase-T will remain an odd niche product. No room for power hogs.”
Then there was, “But I see many customers putting in fiber in the rack in anticipation of 40G/100G.”
To which came the reply, “Correct, why not then 10GbaseT from server LOM [LAN on motherboard] to top of rack switch? 100/1G/10G server LOM big reason 10GT more than niche.”
The discussion continued in similar bursts, and was lively enough to convince me that George Zimmerman was right in more ways than he probably intended. We are sitting at a really interesting time, when twisted-pair copper-based 10 Gigabit Ethernet is about to step up as a bona fide option for many users.
For those of you for whom this is a real consideration, contemplating it will, fortunately or unfortunately, take place in far more complex terms than 140-character bytes.