Slashdot Mirror
IEEE Seeks For Ethernet To 'Go Green'
alphadogg submitted a piece at the NetworkWorld site about the IEEE's efforts to introduce energy efficiency to Ethernet use. The group's Energy Efficient Ethernet group is looking into methods by which standards can be tweaked to encourage power savings. Current plans include ways to make computers 'choosier' about what level of bandwidth they're using. Idle systems would only run at 10Mbps, while email might draw 100Mbs, and scale up to 1000Mbps for large downloads and streaming video. The group is planning to discuss changes to the Ethernet link and higher layers. No restrictions are planned for device manufacturers, although the article suggests some companies might try to use energy efficiency as a competitive advantage. The EEE group estimates some $450 million a year could be saved via the use of energy efficient Ethernet technology.
Seems they are saving energy by throttling bandwidth for the article. Any manage to read it?
The world is made by those who show up for the job.
Once Apple adds the ability to negotiate EEE in Macs, they'll call it iEEE.
"We are all geniuses when we dream"
- E.M. Cioran
Nevermind ... I just had to try connecting 3 or 4 times. Interesting idea. Let's see ... throw out millions of PC's with integrated ethernet, replace them with new machines. Oh, guess they mean in a decade or so through normal replacement.
The world is made by those who show up for the job.
Use more zeros and fewer ones.
One of the easiest ways that the Ethernet people could encourage energy efficiency would be by promoting greater use of Power Over Ethernet. By moving networked devices away from each having an individual wall wart, which are typically inefficient (as well as inconvenient), PoE lets you concentrate the AC to DC conversion in one place, for greater efficiency. As long as you don't have terribly long cable runs, I think there would be a significant net savings overall.
The number of networked devices people are going to have in their homes is only going to grow. I think a big segment could be in "Micro NAS" devices, basically single HD boxes that plug in to a home network and add storage that's accessible from any computer in the home. They're smaller and cheaper than RAIDed NAS solutions, but more convenient for people who have multiple computers than a FireWire or USB2.0 hard drive. And then you have routers, WiFi APs, network cameras, set-top-boxes for playing back video and audio, etc. All of those light-draw devices could be powered over the network connection instead of each having a wall wart.
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
IEEE Seeks For Ethernet To 'Go Green'
That's good because I'm really tired of the white and blue.
Developers: We can use your help.
I did. The problem (FTA):
"One challenge is finding a way to make a PC or laptop network interface card (NIC) change gears more quickly -- "a couple orders of magnitude faster than auto-negotiation, to make the switch as seamless as possible," Bennett says. "Auto-negotiation runs at about 1.4 seconds and we're talking about -- just to start the discussion -- a millisecond of switching time."
So, why not just set NIC(s) to negotiate at the lowest speed first? Then throttle up gradually based on end to end transmission intervals. They talked about using buffers and NIC electrical consumption to handle the negotiation. I say, just start at 10mbps and negotiate up to Gig speed gradually, and make the firmware drivers allow one to turn that feature off/on and back to our current default. My simpleton mind must be overlooking something.
I hope, when they die, cartoon characters have to answer for their sins.
Another suggestion - Stop all the spamming. There must be a coal-powered powerplant's worth of electricity right there.
Take the cheese to sickbay, the doctor should see it as soon as possible - B'Elanna Torres, "Learning Curve"
I smell, DRM, RIAA, MPAA, and a few others behind this and it has NOTHING to do with "energy efficiency". What better way for the *AAs to get more control of your box.
My karma is not a Chameleon.
Sounds like someone is really starting at the wrong end. IMHO.
I'd estimate that power supply inefficiency chews up more than this proposal will ever save. If you spent your time making the power supplies of PC's, Switches, routers more efficient you'd probably have a greater impact. How about better efficiency in the FET's, transistors and amplifier circuitry? Last time I checked, my Ethernet looms didn't get that hot. (isn't it all about "(i^2).R"?. Heck turning off the light in the switch room probably does more to save power. Plus all the heat im my server room is from the servers, not the Ethernet. If your that worried, switch to fiber.
I thought the transfer of data at the physical layer was through the transfer of 'holes' anyway.
Does 100 (or 1000) really take that much power to download one "file", or is it the same amount of power used, just in a shorter time period?
Or is it power used while idle? Does a 1000 device comsume more power idling in that mode than a 10 device would?
Who would win this election: Andrew Weiner vs Andrew Weiner's weiner.
Does that much current actually go over ethernet transmissions? It seems to be that more power could be saved by more efficient power supplies in the switches than by wasting a lot of time and research in figuring a way to throttle link speeds. Does anybody have a value for the amount of electricity used for an hour's worth of data at 10 megabits as opposed to 1 gigabit?
It just surprises me that +/-5 volts over copper really makes all that much difference compared to all the other waste in the datacenter.
Also, what's the difference in energy usage for copper vs fiber links??
...it'll be Windows based only. The non-MS crowd will have to reverse engineer it in a country that doesn't make that illegal.
-"...bad old ideas look confusingly fresh when they are packaged as technology" - Jaron Lanier (Digital Maoism on Edge.o
I see no reason to switch from 568B to 568A.
It would have been nice if the article had broken down the network power consumption down into something useful like the number of Watts for a single 10/100/1000 Mbps port, instead they bury it under the total IT power consumption. In the end, when I have to weigh the cost of upgrading my PC vs the energy savings, I'd like useful benchmarks.
One line blog. I hear that they're called Twitters now.
It seems to me that the energy savings would be more beneficial to whoever pays the bill on the huge server farms rather than individual "normal people" who have a small ethernet running at their house or small business and whatnot. I hope they will shift away from this and focus on another area where they can actually make a difference that would noticeably benefit everyone; I especially like the idea of improving power supply efficiency (which is a bigger problem that just ethernets, IMO). One way to do this would be to get devices running directly off of direct AC current. IIRC, you can avoid a 20% penalty hit incurred from during the conversion from AC->DC that you get with normal power supplies.
:p
Beware of the gloves of the over-complicators
--
Wi-Fizzle Research
Censorship is obscene. Patriotism is bigotry. Faith is a vice. Slashdot 2.0 sucks.
How does a typical gigabit ethernet cabling installation compare to the same network using fiber cabling in terms of power usage?
...is this group led by ethernet equipment vendors? Perhaps vendors who are unhappy with the recent decline in equipment upgrades since people aren't upgrading from gigabit or even to gigabit from 100mbit in a way that helps their stock price sufficiently?
It seems to me that, considering the number of ports active out there, they're talking about a tiny amount of savings per port for a total investment that could have a much larger effect if spent elsewhere.
Hell, I bet more power is wasted by the power supplies, overly conservative fan controls, uncleaned air filters, shorted out UPS batteries that should have been replaced decades ago, overpowered CPUs, and crappily written firmware of the currently deployed switches than is consumed by transmission losses.
IEEE-EEE?
eye sextuple E eye? Or did a joke just whoosh over me?
Imagine fiber with green laser - how green is that!
Yesterday was the time to do it right. Are we having a REVOLUTION yet?
For broadcom Ethernet PHY chips, they use about 1W/port when configured as 1000BaseT (GigE). GigE require some heavy duty DSP filtering as well as driving 4 pairs of bidirectional transceiver. They would burn less power when they are running at 100BaseT which only to drive 1 pair of receive and transmit. Not sure if there are significant saving going down to 10BaseT as the number of transmit pairs and the DSP's are dominant.
While this might not seem a whole lot of power, when you are looking at Enterprise size (say a few hundreds to thousands ports) setup, there can potentially be savings at the few hundred watts to thousand watts range.
At an office I once worked, there were a lot of spare switches laying about after upgrading to 1000BaseT. They were considered "spare" or whatever, but there was a great many... so I sorta brought one home and mounted it into my rack and used it for a couple of months. The next two electric bills made me rethink how nice it looked to have a 24 port switch in my rack instead of that cheapy 8 port sitting on a shelf. It consumed a NOTABLE amount of power. Now, there were other things involved I'm sure... things like the changes of the seasons, global warming and all that. But when I brought the switch back to the office and went back to my cheapy 8 port again, I saw a change in my power bill.
If I ever decide to spend money on a nice looking switch, I'll be sure to reference the power draw of the units I review.
Meanwhile, far more than $450 million would be spent on IT support services, troubleshooting problems created by computers that keep changing their link speed.
Tired of FB/Google censorship? Visit UNCENSORED!
"There's lots to take on with this effort," says Mike Bennett, senior network engineer at Lawrence Berkeley National Lab, and chair of the EEE Study Group.
Uh huh. Tell me how much a 1000 vs 100 vs 10 port uses up in electricity when IDLE, until that number shows to be wildly different this is nothin more than a scam to get more money for questionable research.
I have no idea where the savings could come from. They shouldlook into spreading more efficient caching mechanisms so that people aren't waiting for webpages to download, thus reducing their time at the computer.
Wow, that's awesome! Now when you get DDoSed and your power suddenly spikes above the level it's been the last three years, you start popping breakers!
So does this mean that the internet, created by Al Gore is in some way responsible for global warming? If we didn't have all of these computers accessing networks, forcing us to burn fossil fuels for power then maybe we would be about 1C cooler. Thanks Al - for everything!
I think the problem is that they don't want the throttling up to be gradual, they want it to change gears quick like.
A blog about stuff.
Yeah, I used to have a catalyst 5000 with two 12 port cards and two 48 port 10mbps cards. But I didn't want to pay the power bill so I sold it. Now I have three tiny 10 port 10Mbps switches around the house doing switching things. Sure, I don't get any management, or vlans, or what have you, but I'm not sucking down an amp just running cooling fans, either, let alone what it takes to run one of those bastards. And that's not even that big a switch!
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
So a straight-through is: Green white / Green / Green white / Green / Green white / Green / Green white / Green Sweet. Crossovers then would be: Green white / Green / Green white / Green white / Green / Green / Green / Green white So much easier to remember! Thanks, IEEE!
> idle or underutilized Ethernet connections more energy efficient
:)
There are several ways to increase measured efficiency. Two of them include:
1) Load the network with verbose transmission protocols, junk, or spam such that more network cards have higher sustained traffic (quantity means more than quality from the usage point of view).
2) Increase the number of hardware exploits such that underused network adapters can be continually used by those who know of the hardware exploits (make the network adapters available to those who have convinced themselves that they need more bandwidth than they're willing to pay for)
This is _not_ a troll.
the NPG electrode was replaced with carbon blac
DC-DC converters are fairly efficient and it doesn't really matter if it was on the mobo or a separate unit, you'd still need it. So you have a 12v (or maybe higher since there is less loss) supply all over and then have little DC PSUs inside your PCs.
- it's the power required to process the packets. More or less, a GigE card should need 10X (divided by some fudge factor that probably makes the real ratio closer to 2 or 3X) the compute power of a 100Mbit card. Processing GigE at full throttle actually takes quite a bit of CPU - we don't notice it much because most GigE interfaces have a TCP Offload Engine that avoids bogging down the CPU and bus.
So your TOE could easily have a variable speed CPU that basically goes to sleep when it can negotiate the physical interface down to 10Mbit, or whatever. SOunds pretty straightforward.
Give a man a fish and you have fed him for today. Teach a man to fish, and he'll say "WHERE'S MY FISH, YOU IDIOT?"
Okay, there's two things that could consume power in an ethernet connection: maintaining the circuit at a given voltage level (ie steady state), and switching the signal (ie, sending a bit, modulo whatever encoding scheme is used).
For a given chunk of data -- email, text message, video clip, whatever -- you have a certain number of bits which is going to require a certain number of signal switches, whether you do those at 10M/sec or 1000M/sec. So, no energy savings there.
You also need to keep the circuit open, regardless of whether or not you're sending a signal over it. (Unless you want to try syncronizing the connection times: "call me every 7 minutes past the hour" or something. Not very practical). No energy savings there either.
So how, exactly, do they figure this is energy saving? (If anything, the extra bits sent for negotiation make it worse, no?)
-- Alastair
I agree. I guess I was thinking more along the lines of how your turbo charger works in a vehicle. There is that slight delay before it kicks in, but not very perceptible overall.
I hope, when they die, cartoon characters have to answer for their sins.
It's the square of the clock speed; it comes from some math in second- or third-year Electrical Engineering.
Unfortunately it may be a bit late for this. Modern ICs have such small features, and electrons are such large, fuzzy objects, that leakage current has become large. In the generation being used for current designs it amounts to half the power consumption. Leakage doesn't change with speed - or even if the clocks are actually stopped! You have to turn the power completely off to to some chunk of the chip if you want to reduce it.
Unless/until chip manufacturers find a way to reduce leakage again and deploy processes using it, power benefits from slowing clock speeds are likely to be too small to be of interest.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Thanks. I haven't had as good a chuckle in some time.
... ... collisions...
Those who do not understand autonegotiation are doomed to repeat... packets... thanks... to...
(Sure, there are some cases of crap out there, but I've spent ***much*** more time fixing problems caused by not understanding how negotiation works.)
There are a lot of solid-state DC to DC converters available today. Zeners are probably the easiest to understand conceptually, but there are discrete chips that you can run out and purchase today that will take a wide range of DC input voltages, and put out and lower voltage that you want. Some of them are fairly efficient, too. (I'm not sure how efficient they are compared to using discrete Zeners.)
There are even ones that will go from a low voltage to a higher DC voltage (obviously drawing more current from the supply side, they don't break any laws of physics). I think they do this by converting to high-frequency AC internally, and then using a very small switching power supply to produce the desired output voltage. But the power they produce is pretty clean.
I've been told that modern motherboards have quite a variety of DC-to-DC converters on them as it is today, so it really wouldn't be that hard to build a board that ran off of a single supply voltage, and then produced whatever other voltages it needed using its own power circuitry. Obviously you'd want to reduce the number of different voltages (to keep parts count down, particularly of heat-producing converters), but there's no new technology required there. Getting 3.3V, or whatever the RAM and PCI require, or 5V for USB, from 12V is pretty simple; it's mostly an exercise in optimization for cost or power consumption.
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
Stop all the downloading!
I'm a computer.
Don't get me wrong. I'm all for being green. But it would seem that instead of putting all of that effort, design time, and eventual costs in equipment in order to save a very small number of watts on the ethernet chips at each end of the link, a slightly larger effort directed into power supply losses, CPU power usage, or GPU power usage would yield 10x the benefits.
Realistically, I know that they can't just walk over to Intel, AMD, and NVidia, and say "Alright, guys, we're here to tell you how to use less power." They're just doing what they can, and they deserve applause for it.
Oh, you're not stuck, you're just unable to let go of the onion rings.
I don't know how much current you can source with such a design... I suppose with some power MOSFETs in parallel and some big-ass filtering caps you could get the desired effect.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
100Mbps for email!?!?!
Let's say an typical e-mail is 10KB, so that's 80Kb.
100 * 1000 / 80 = 1250 e-mail messages per second. I don't know anyone who can type that fast. Maybe they meant e-mails with video attachments or something like that.
Even if it's a 5MB e-mail, you could still send 2 per second at 100Mbps.
I've read all the posts and this is the first one that makes any sense at all to me.
10 Base Ethernet uses about the same amount of power as 100 Base. There's power required to keep the loop active, and there are data transitions. The data transitions are not compressed with 10 but they are with 100 (and moreso with 1000). I could see how 1G Ethernet would use more power due to the extra loops being active -- but not very many folks actually use 1G Ethernet.
To me this whole thing seems like an early April Fools joke.
JSL
I'm a Hardware Design Engineer that has done a zillion ethernet designs.
The reason 1000Mbps consumes more power than say 10Mbps is because of all the DSP and signal transitions that the physical layer chips have to do to work at 1000Mbps over 100m of cable compared to 10Mbps. This will become more and more of an issue with 10Gbps ethernet.
With respect to PoE, in practice the waste of power by sending lower voltage over 24 Gauge wires is a essentially wash when you compare it to the inefficent wall-wart power it replaces because those transformers are much less effecient than the flyback transformer in the powered device. When you look at total power loss it is about the same: 60% efficient.
Given the number of times that autonegotiation has given me headaches because supposedly compliant devices couldn't agree on how to setup a connection, I wouldn't want to set this up on any of my networks. I just can't see myself explaining to the CIO that the reason that the ERP is slow to the point of being unusable is because the core switches renegotiated their bandwidth down to 10Mbit/sec overnight when they were unused and were unable to ramp it back up again correctly. There is a reason that autonegotiation is often disabled & it's called experience...
Democracy is a sheep and two wolves deciding what to have for lunch. Freedom is a well armed sheep contesting the issue
There is a reason that autonegotiation is often disabled & it's called experience...
Usually it's called resistance to change. I haven't seen any trouble since around 2000. Except when the idiot at the other end of the cable locked it at 100-full, forcing my end to go 100-half. Luckily that problem is gone with gigabit, since that is autonegotiation or nothing.
Finally! A year of moderation! Ready for 2019?
I spent part of the holidays on a transatlantic trip to debug a network where applications being used to keep track of trains in a subway were failing. Periodic misnegotiation of the Ethernet parameters was a major part of the problem that disappeared once the ports were set statically. Part of the problem was that some of the equipment was a few years old, but then we didn't have the luxury of telling the client that all he needed to do to have a functional network was to replace the oldest 20%.
If you have the luxury of being in network environments where autonegotiation just works all the time, well lucky you. The experience of my peers has been that if you need it to work, turn autoneg off. If you don't care because it's just a lambda PC being connected to the link, leave it on.
Democracy is a sheep and two wolves deciding what to have for lunch. Freedom is a well armed sheep contesting the issue
$450 million dollars! I really hope my boss reads this! I can just see his beady little eyes all a-glow at Monday morning's staff meeting. It'll be a green glow alright!
This means that we in the IT department will be given another measure of job security, or fired, because nothing will work and we'll have to buy new equipment to comply with the new standards. You know we will.
The IEEE didn't do this just because it means selling more gear, but I'm sure they weren't really upset at the idea either. Let's not delude ourselves too much about that point, it's not as though the members of the IEEE have nothing to do with the business of selling networking equipment, although I'm sure their first concern was the savings they could pass on to thier customers. They're propably really concerned about the environment too, I guess.
There are only two cases I know of where autonegotiation will cause failures:
a) You are using shitty wiring not capable of the maximum common speed of both endpoints. (i.e. wiring only capable of 10M, endpoints can both do 100) Unfortunately autonegotiation only deals with endpoint capabilities, not wiring.
b) You have devices that are in forced 100M full duplex. Forcing 100M full duplex should only be done by a device if the forcing is done by someone who knows EXACTLY what they're doing, as combining an autonegotiating endpoint with a forced full duplex endpoint results in a duplex mismatch. The IEEE standards explicitly state that 100M full duplex should never be used with an autonegotiating endpoint, and in fact forcing full duplex should be avoided in nearly all situations.
That said - the power differences between the various Ethernet modes are negligible. The common aspects of all modes (MAC/PHY chipsets, the computer doing the communications) are going to be FAR more than the difference in power consumption between the modes. There is little to no benefit to trying to "throttle back" Ethernet, unless they intend to create entirely new 10M and 100M modes. (If you applied all of the new signal processing algorithms used for GigE and reduced the clock rates, you could also reduce the output voltage swing, and as a result save power.)
retrorocket.o not found, launch anyway?
My guess is that the new gear you brought home was a managed switch. Rather than having an ASIC dedicated solely to basic Ethernet switching, managed switches typically have a similar ASIC coupled with a general purpose CPU that controls the thing.
The per-port power (especially for idle unconnected ports) was probably negligible compared to the difference between managed "big iron" Ethernet switches and a basic "dumb" switch.
retrorocket.o not found, launch anyway?
Stop using shitty wiring (Autonegotiation only deals with endpoint capabilities, not with cable quality), stop connecting forced full duplex equipment to autonegotiated equipment (why? go read the 802.3 spec or just take my word that it isn't going to work well).
Problem solved.
retrorocket.o not found, launch anyway?
There was an ambiguity in the earlier versions of the autonegotiation specification, that caused earlier devices to be incompatible. In particular, Cisco got in wrong (or at least their interpretation was disallowed when the specification was refined).
So, nowdays I would say that far more problems are caused by people forcing full duplex on one end, when autonegotiation would work fine. Yes, if you hit some old kit you might be unlucky but the world would be a better place if we put an end to this "autonegotiation is unreliable" myth; even if it happens to be true on occasion.