| At a Glance | |
|---|---|
| Product | D-Link "Green Ethernet" 8-Port 10/100/1000 Desktop Switch (DGS-2208) |
| Summary | 8-port unmanaged gigabit switch with power-saving technology |
| Pros | • Jumbo frame support to 9K |
| Cons | • No measureable power savings |
Yesterday was Earth Day, that day of the year when companies remind you that they really do care about the Environment (really!). So in the spirit of the day, before I make my weekly visit to my county recycling center, I’ll take a quick look at a product that’s intended to help your small network to save the planet.
D-Link actually introduced the "green" versions of the DGS-2205 and DGS-2208 in October 2007, then added power-saving versions of the DGS-1016D 16-port and DGS-1024 24-port unmanaged gigabit switches in February. And, as I was writing this, it looks like D-Link has added three "smart" switches to the "green" lineup for a total of seven.
D-Link says that the power savings are achieved primarily through two power management techniques. The first adjusts power to the Ethernet port drivers based on cable length. The second automatically detects when devices attached to each switch port are powered down and puts the corresponding switch ports into a reduced-power standby mode.
A cute graphic on the 2208’s web page claims up to 80% power savings. So, ever the skeptic, I asked D-Link to send me a Rev C1 2208 so that I could test the claims.
According to an email exchange I had with D-Link, the "secret sauce" inside at least the DGS-2208 is a Vitesse VSC7398 SparX-G8e Enhanced 8-port Integrated Gigabit Ethernet Switch with Transceivers. But when I opened up the Rev C1 DGS-2208 that D-Link sent for review to snap the photo below (Figure 1), I found a VSC7388 SparX-G8 8-port Integrated Gigabit Ethernet Switch.
Figure 1: DGS-2208 board
According to the "Energy Efficient Solutions for Ethernet Electronics" product brief downloadable from both the VSC7388 and VSC7398 pages on Vitesse’s website, "Vitesse’s SimpliPHY™, SparX™, and G-RocX™ devices incorporate two advanced green technologies that can reduce power well over 400 mW per port, or up to 80% compared to active 1000BASE-T power consumption".
The brief then goes on to describe two mechanisms that are essentially the same as those that D-Link describes on its "Green Ethernet" product pages. (Vitesse has dubbed this "EcoEthernet".) So it appears that D-Link didn’t even need to change devices to get the "Green Ethernet" features; just wait for a new revision of the same device that it had been using.
Does It Work?
I set up a simple test to check D-Link’s claims. I didn’t have a previous-revision DGS-2208 for comparison, so I grabbed a Trendnet TEG-S8 that I had left over from the 8 Port Gigabit Switch Roundup. The TEG-S8, like most of the other gigabit switches in the roundup uses the Broadcom BCM5398.
I then rounded up eight devices that included three desktop computers (each with a gigabit Ethernet NIC) and five devices with 10/100 Ethernet (three notebooks, one NAS and one 10/100 switch uplink port). I plugged the switch under test’s wall-wart into my trusty P3 Model P4400 Kill-A-Watt power meter and took a Watt reading with no devices plugged in.
I then plugged in all devices but with power off on each one and took a second Watt reading. Finally, I powered up all devices, waited for them to finish booting and took a final power reading. The results are presented in Table 1.
| D-Link DGS-2208 |
Trendnet TEG-S8 | |
|---|---|---|
| Unplugged | 1 W | 2 W |
| Plugged, powered down | 2 W | 2 W |
| Plugged, powered up | 5 W | 5-6 W |
Table 1: Power Consumption Comparison
As you can see, both switches behaved similarly in terms of power consumption. The small differences you see are more likely due to line voltage variation and rounding in the Kill-A-Watt than any real difference in switch power consumption.
I couldn’t find any claims by Broadcom for power-saving technology in the BCM5398, by the way. So I don’t know why the two products using different devices behaved essentially the same in terms of power consumption.
I didn’t attempt to verify the line-length power-saving claim, since I don’t have that much cable lying around. But even if I did, I don’t think the outcome would be any different.
I’m all for doing what we can to reduce energy consumption and wasted energy. But you’ll save more power by shutting off your PC and monitor when you’re done with them for the day, than you’ll get by using D-Link’s "Green Ethernet" or any manufacturer’s small gigabit switch.