Facebook Letting Everyone See How Much Data-Center Power It Consumes 86
Nerval's Lobster writes "Facebook has added real-time dashboards for measuring the efficiency of its data centers' internal power and water use. Two dashboards monitor the company's Prineville, Ore. (here) and Forest City, N.C. data centers (here), measuring both the Power Usage Effectiveness (PUE) and Water Usage Effectiveness of those facilities, in addition to the ambient temperature and humidity. So far, visitors to the Prineville and Forest City dashboards only see a limited snapshot of the Facebook data: the display only covers 24 hours, and is delayed by 2.5 hours on both sites. Facebook also hasn't disclosed how many servers the data represents, which could conceivably be used by competitors to get a sense of the social network's total computing power. The company said that once its data center in Luleå, Sweden, comes online, Facebook will begin adding data from that location, as well. Although Facebook said it provided the information out of a sense of openness, the data—showing PUEs of about 1.09 for both facilities as of press time—is a bit of a boast, as well; as recently as 2011, Uptime Institute said that the average data center's PUE was approximately 1.8. So far, Facebook hasn't said whether it will provide access to the dashboards via an API, so third parties can get a better sense of how Facebook is managing power and water use over time, and through various seasons of the year."
Number of Servers (Score:5, Funny)
Makes perfect sense they don't wanna disclose the number of servers. They like their privacy
Privacy dashboard (Score:5, Interesting)
.
- what data nuggets have been collected about me over the past 24 hours, week, month
- what third party entities my data has been shared with
I am sure that this community can suggest other items that would be useful on a Facebook Privacy Dashboard.
In the background, I cannot shake the thought that Facebook is putting up this energy consumption dashboard for the purpose to divert attention away from Facebook's ongoing privacy issues.
Re:Privacy dashboard (Score:5, Insightful)
Will deleting change anything? (Score:3)
Delete your account, stop worrying, and get some sleep.
Its off topic (but a more interesting topic) Although having read about it you have to *delete* the account which allegedly will remove it, but deactivate your account and messages you sent, may still be visible to others. they also save your timeline information (ex: friends, photos, interests, etc.)
But even then you have companies like http://www.guardian.co.uk/world/2013/feb/10/software-tracks-social-media-defence [guardian.co.uk] like this one who mine data...and create links. Not sure how you delete their data, or even
Re:Will deleting change anything? (Score:5, Interesting)
if you change you age to under 12 facebook nukes you account from orbit or so i am given to understand.
Re: (Score:1)
You can't change your Facebook age to below twelve if you never were on Facebook.
Re: (Score:2)
If you never opened a Facebook account, then Facebook CANNOT have violated your privacy. There is no way that Facebook is running bots collecting data from everyone's computer and such. That means that if you never told Facebook about yourself, then it was your friends and family that did so and they are the ones that violated your privacy, not Facebook.
Re: (Score:1)
Re: (Score:1)
Re: (Score:1)
Even not having an account in the first place won't stop them. Every time you see a facebook 'like' button on a webpage, Facebook knows where you are. They build web history profiles even of non-users. It doesn't give them nearly so much data to process as someone with an account could, but they do what they can with it.
Re: (Score:1)
Use Ghostery [ghostery.com] to escape this.
Re: (Score:3)
I cannot shake the thought that Facebook is putting up this energy consumption dashboard for the purpose to divert attention away from Facebook's ongoing privacy issues
Expect advertisements for psychological counseling the next you log in.
Deceptive metrics (Score:5, Insightful)
Of course, "Power Usage Effectiveness" and "Water Usage Effectiveness" are somewhat deceptive metrics, because there's little useful societal "effect" produced by running Facebook's massive spyware operation. No matter how efficiently they churn out clock cycles per kWh or liter, spending those clock cycles on Facebook is an ecologically disastrous misapplication of humankind's resources. There is nothing "effective" about growing the share of the economy devoted to advertising.
Re:Deceptive metrics (Score:5, Insightful)
And spending time on Slashdot is a good use of humankind's resources? What is the point of even being human if we can't have a little fun every once in a while?
Re: (Score:2)
I like having fun, and consider it a worthwhile use of power and water --- but I'm pretty sure there are more "effective" ways to have fun with less burden of "creepy stalker megacorporation inserting themselves into the entire fabric of your life."
Re: (Score:1)
I'm pretty sure there are more "effective" ways to have fun with less burden of "creepy stalker megacorporation inserting themselves into the entire fabric of your life."
Sent from your iPhone...
Re:Deceptive metrics (Score:5, Interesting)
Re:Deceptive metrics (Score:5, Insightful)
Like you say, I might hate how Facebook does things, which is exactly what I'm doing here. New forms for fluid interpersonal communications via network channels? A-OK with me. Monetizing and commoditizing the "social network connectivity graph" to further entrench corporate power at the fundamental level of interpersonal interactions? DIAF.
Re: (Score:3)
We havent even begun to understand the positive or negative effects of the communication level Facebook allows. Like it or not, it s a HUGE hive of human activity engaging more people then any other service before. It represents hopes and fears and dreams and thoughts and loves and hates. You might hate how Facebook does things, but the actual events occurring inside are meaningful and useful to the human race.
I've got to problem with facebook. I don't use them though, because If anyone's monetizing my data, I want a cut. Also, every service they provide I've been using since before the Internet. Hell, my BBS had packet radio via my HAM setup -- That's country-wide mesh networking without wires. Facebook is simply the AOL of websites. It's popular and helps grandma use high tech non-features, but it's still shite to anyone with a clue.
Re: (Score:2)
Protests against the government of Argentina last night, largely organized via Facebook and other social networks:
http://clarincomhd.tumblr.com/image/48316699321 [tumblr.com]
Does this give you a sense of scale, about the level of communication people can have on Facebook?
Re: (Score:2)
Strangely, mass protests are documented to have existed in history well before Facebook arrived --- however did they do it? Indeed, Facebook served as a communication channel for facilitating this protest. However, Facebook's ability to track, record, analyze, and sell the participants' data did not help. Nor would a profiteering megacorporation (with its own friendly FBI ties) be an advantageous ally in organizing protests against the powers closer to Facebook's empire.
Re: (Score:2)
I think mass protests were largely local events. The 19 and 20 of december of 2001 we had massive protests (much smaller than the current ones) that ended with then-president De La Rua resigning. Those protests were held in Buenos Aires and only 1 channel was showing them until people called the other news networks about how big it was and they were shit because they weren't showing them. Hours later all channels were covering these events. Once that happened, the protests replicated all over the country, b
Re: (Score:2)
Nice troll, but most people here are familiar only with facebook, and the odd one who uses Twitter.
Re: (Score:2)
No, they haven't. Very few people used a dating site called "badoo", and the first relevant social network was Facebook. Myspace wasn't even a thing over here.
Re: (Score:3)
Advertising keeps the economy running. If everyone stopped buying pointless crap they don't really need or want, we'd see mass-unemployment in a year.
Re: (Score:2)
People don't need 20% of the entire economy (rough estimate, hard to make precise; there is more cost to advertising than direct TV/radio/Facebook expenditures; there is also wasteful effort put into packaging, just-for-selling-points R&D, higher prices for inferior goods with better advertising, monopolization and barrier to entry in markets, etc.) dedicated to "making them buy shiny new stuff" --- they'd still buy plenty if they could. We already see mass unemployment, because the middle class doesn't
Re: (Score:1)
Which companies are "making" you buy their products?
I know mine is an unpopular opinion, but corporations are groups of people united for the purpose of producing and selling. People have the right (at least in the US) to speak freely and that includes enticing others to buy their products. We generally refer to that as advertising.
Would you rather the government controlled every bit of information you are allowed to see?
Re: (Score:2)
Pay attention to the thread above; my statement about "making you buy stuff" wasn't a complaint that some company is forcing me to buy stuff through advertising --- it was a response to the parent poster above claiming people would magically stop wanting to buy things if advertising was cut back. Thus, my point was that companies don't make people buy things with advertising --- they'd be buying things anyway.
As for corporate free speech in advertising: I support free speech. That doesn't mean I support wha
Where? (Score:2)
The amount of data center power consumed would sure be an interesting statistic, but it isn't anywhere in the link. The only two metrics listed other than the publicly available weather are PUE = [Total Building Load (kWh)] / [IT Load (kWh)] and WUE = [Volume of water required to condition data hall air (liters)] / [IT load (kWh)].
Not real time (Score:2)
The dashboard is not actually in real-time, but carries a 2.5 hour delay.
Re: (Score:2)
The dashboard is not actually in real-time, but carries a 2.5 hour delay.
That's it! I want my money back!!!
What a strange world we live in... (Score:5, Insightful)
...where a giant company worth billions--just because people in suits say so--is building state-of-the-art data centers around the globe to store crappy photos of mundane activities and asinine conversations about nothing in order to collect data on consumers for advertisers so they can sell them more gadgets to take even crappier photos of even more mundane activities. (And yes, I'm aware of the irony of appearing on television in order to decry it, so don't bother pointing that out.) Meanwhile the funding agencies that drive the creation of all this technology are being gutted to shave a few fractions of a percent off of the federal budget, Wikipedia is begging users for cash, and NASA had to scrap its shuttle program. Our priorities are a joke.
Re: (Score:3)
...where a giant company worth billions--just because people in suits say so--is building state-of-the-art data centers around the globe to store crappy photos of mundane activities and asinine conversations about nothing in order to collect data on consumers for advertisers so they can sell them more gadgets to take even crappier photos of even more mundane activities. (And yes, I'm aware of the irony of appearing on television in order to decry it, so don't bother pointing that out.) Meanwhile the funding agencies that drive the creation of all this technology are being gutted to shave a few fractions of a percent off of the federal budget, Wikipedia is begging users for cash, and NASA had to scrap its shuttle program. Our priorities are a joke.
Spot on.
Re: (Score:2)
Well I'm certainly not going to pay enough for it to be HBO.
Re: (Score:1)
Re: (Score:3)
Re: (Score:2)
the majority of people in the WORLD doesn't use facebook (alleged: 1 billion people out of 7 billion)
Facebook: a Network Effect virus (Score:2)
.
[In the middle of writing this post, my SO wanted me to "unfriend" some supreme PoS. No visible way to do this. Search Google. Go to this Facebook help page [facebook.com], appropriately called "How do I unfriend or remove a friend?". In attempting to follow the 3 simple steps, I go to that person's Facebook page and attempt t
Tagged "whocares" (Score:3)
I take an extremely accepting view of what might qualify as "news for nerds," but this absolutely fails the "stuff that matters" test. Honestly, who the hell cares about this? It's a cheap stunt, and nothing more.
Re: (Score:3)
I take an extremely accepting view of what might qualify as "news for nerds," but this absolutely fails the "stuff that matters" test.
As some other commentator noted, those phrases appear to be being removed from the Slashdot site. Hit CTRL+F and try and find them.
Re: (Score:2)
Fair enough. I suppose I've just become so accustomed to the presence of those words (since about 2000 I guess) that I missed their absence. I'll continue to lower my expectations accordingly.
Re: (Score:2)
Re: (Score:2)
The text persists in the head within a link title:
But it does indeed seem to be gone from any normally viewable place, sadly.
Re: (Score:2)
<title>Slashdot: News for nerds, stuff that matters</title>
However, since many browsers, especially those on Windows, dropped the title bar for more viewable screen area, it's often not shown. It does flash up for a few hundred milliseconds in the tab text in FF on Windows but is rapidly replaced with "Slashdot".
Re: (Score:2)
You're quite right. I was only looking in the HTML for this thread, but it's still in the title for the the main homepage.
So timely! (Score:1)
I, for one, was up ALL NIGHT worrying about how much power they were consuming, and how efficient they were. I was just getting ready to write my congressman about it. Facebook really anticipated my concern here! it's like they read my mind, or my email!
Uh oh.
Here's wondering... (Score:2)
How much extra power is this going to use?
Here (Score:2)
Two dashboards monitor the company's Prineville, Ore. (here [facebook.com]) and Forest City, N.C. data centers (here [facebook.com])
Why add a separate word "here" just for the link? That part could have been written like:
Two dashboards monitor the company's Prineville, Ore. [facebook.com] and Forest City, N.C. [facebook.com] data centers
Much neater.
Doesn't show power consumption (Score:4, Informative)
This doesn't show power consumption. It only shows ratios that are considered a sort of measure of efficiency.
It's like showing "miles per gallon" instead of "gallons used". In the case of facebook, they may be driving at 40 MPG, but they drive a million miles a day and that's a lot of fuel!
U.K. National Grid Status (Score:2)
Bah! Just the power consumption of some uninteresting company, here is the real time power consumption of an entire country [templar.co.uk].
Terrible Visualization (Score:3)
On the other hand, they failed at one of the axioms of data presentation: they didn't provide scale for their axes. The human eye/brain isn't that good at judging radial amplitude, just like it isn't good at discerning logarithmic amplitude (which is why we have log plots: to linearize it). Down in the corner they mention that the circle represents the past 24 hours, but they aren't any graduations on the graph (e.g., 1-hour tick marks). Because the graph represents 24 hours instead of 12, our usual sense of time:angle from analog clocks is off by a factor of two. If you look at it long enough, you can work it out, but a good data representation shouldn't require that. If you hover over a particular measure (e.g., PUE), it'll hide the other traces (a nice touch, perhaps), and will show you the scale minimum and maximum. But, again, because it is a polar plot without gridlines, it's damn near impossible to read and figure out, say, what the PUE was 5 hours ago.
Oh, but wait, they added a cursor, so that you can roll it back to a certain time and get the values. How very clever! I'll bet the 20-year old intern that implemented that got an awesome pat on the back and course credit for industrial design. But it doesn't negate the fact that a good data visualization should be self-evident: you look at it and immediate see what's going on. You shouldn't need to "query" the graph by interacting with it; it should stand alone.
Would an ordinary X-Y plot, with gridlines, really have been that difficult, or cramped their precious design that much?
Re: (Score:2)
Actually, I thought it was quite clever.
You must be a senior, or at least act like you're a senior because you can't stand anything thats different.
It took me about 3.2 seconds to understand how it works.
Re: (Score:1)
I guess you've never seen a chart recorder? There was one on the wall in the lab where I worked (many a year ago) at NCSU -- used to be a thermal test lab. We also had them all over the place in the hatchery my grandparents ran (many decades ago.) The rotating chart is a very efficient means of recording one or more readings continuously without using a huge roll of paper.
(google "temperature chart recorder")
Re: (Score:2)
I am familiar with those (see the last sentence of my first paragraph). Yes, they are efficient in terms of the resulting record (a disc of paper with lots of data on it) and the mechanism (a pair of motors), but few would claim they are all that good in terms of readability. And, I'll note, they all have gridlines printed on them, so that you can actually read the data afterwar