Talk:Utah Data Center

Data Centre

This is text from the Camp Williams artcile that was deleted (I have restored) but has several older refs re the 'Utah Data Centre', which was apparently previously called Community Comprehensive National Cybersecurity Initiative Data Center.

220 of ^Borg 10:24, 9 March 2014 (UTC) nb. Restored text from April 2012

==Data Center==

The National Security Agency (NSA) is building the $1.5 billion Community Comprehensive National Cybersecurity Initiative Data Center at Camp Williams, the first in a series of data centers required for the Comprehensive National Cybersecurity Initiative. The 1.5 million square foot facility built on 200 acres is expected to store 1 yottabyte of data by 2015. The facility which is expected to be completed in October 2013 will use 65 megawatts and will cost another $2 billion for hardware, software, and maintenance.

- 220 of ^Borg 04:52, 21 April 2012 (UTC)

Good data. Added.  Done--Tomwsulcer (talk) 21:13, 25 August 2012 (UTC)

About the addition

About this added paragraph:--Tomwsulcer (talk) 18:35, 27 August 2012 (UTC)

On August 22, 2012 The New York Times published an Op-doc in a forum of short documentaries produced by independent filmmakers that was produced by Laura Poitras and entitled, The Program.[9] It is a preliminary work that will be part of a documentary planned for release in 2013. The documentary is based on interviews with William Binney, a 32-year veteran of the United States National Security Agency, who became a whistleblower and described the details of the Stellar Wind project that he helped to design. The program he worked on had been designed for foreign espionage, but in 2001 was converted to spying on citizens in the United States, prompting disclosures by him and others concerned that the actions were illegal and unconstitutional. The subject implies that the facility being built at Bluffdale is a facility that is part of that domestic surveillance, intended for storage of massive amounts of data collected from a broad range of communications that may be mined readily for intelligence without warrants.[10] The author reports that on October 29, 2012 the United States Supreme Court will hear arguments regarding the constitutionality of the amendments to the Foreign Intelligence Surveillance Act that were used to authorize the creation of such facilities and justify such actions.

My concerns are regarding WP:OR and WP:UNDUE. Can the new paragraph addition be shortened to only those aspects that explicitly refer to the Utah Data Center?--Tomwsulcer (talk) 21:00, 25 August 2012 (UTC)

'Bumblehive'

http://www.dbia-mar.org/downloads/Army%20DBIA%20Mar16%202010.pdf references Army Core of Engineers project 21078 in a report from March 16, 2012, listing the project as $ 1,489,000,000 and calling it "Bumblehive UDC (Incr 1- 5)" - UDC probably stands for Utility Data Center.67.188.202.139 (talk) 20:34, 2 December 2012 (UTC)

Yottabytes?

Multiple yotabytes? Absurd. That exceeds the worldwide platter production by a an order of magnitude (annual production runs around 500M). At 300Mb/sec, it would take 10 million years to write that much data. (yes, I recognize this would be done in parallel, I'm trying to put a scope on the problem). On tape, the only feasible solution (due to power/heat issues), it would take more in the 100M year range. Multiple drives would be failing every second. Come on, this is supposed to be an Encyclopedia.

All of these stories come from the same CNET source, which misread a statement that we need to eventually handle a yottabyte of information. Handle, not store. I quote the relevant information "as a 2007 Department of Defense report puts it, the Pentagon is attempting to expand its worldwide communications network, known as the Global Information Grid, to handle yottabytes (1024 bytes) of data." That's a world wide network, not the Utah facility, and the word used is 'handle'.

I am editing the article to remove the yottabyte claim.

  — Preceding unsigned comment added by 71.92.252.196 (talk) 17:11, 14 June 2013 (UTC) 

I agree, the yottabyte claim is completely absurd. I did some calculations on this. A yottabyte would require a data center would cost $10 Trillion a 1 cent per GB, almost the whole GDP of the united states. And judging from the density of magnetic storage medium, it would also have to be 15 miles across, which is clearly counter indicated by the photons of it. Here are my calculations. Pulu (talk) 21:58, 21 July 2013 (UTC)

Surely the article is in error. 1 Yottabyte = 1 Trillion Terabytes. At a price of $50 a terabyte (very conservative estimate) this would put the cost of the data center at $50 trillion dollars which far exceeds the federal budget of 3.6 trillion. — Preceding unsigned comment added by 71.201.194.191 (talk) 22:03, 17 April 2013 (UTC)

From the Wired piece: It needs that capacity because, according to a recent report by Cisco, global Internet traffic will quadruple from 2010 to 2015, reaching 966 exabytes per year. (A million exabytes equal a yottabyte.)

That's 1 one-thousandth (0.001) of a yottabyte in a year. Even if you assume 10 times that, it'll take a long time to get to a yottabyte. (Of course, there's legacy data to consider, too.)

Current estimates have the center requiring $2B/year for hardware, software and maintenance. My guess is that $50/TB is way too high. This is the federal government we're talking about here. Woodshed (talk) 00:10, 18 April 2013 (UTC)

"capable of storing" can mean the complex is capable of that much storage after they add enough drives, thus it might mean they currently don't have that much storage yet. • Sbmeirow • Talk • 03:49, 7 June 2013 (UTC)

Regarding cost, remember that the buyer is not your ordinary consumer! — Charles Edwin Shipp (talk) 14:12, 8 June 2013 (UTC)

Even so, it would still cost an obscene amount of money. If you use the current max capacity drives on the market, 4 terabyte drives, then it would take 250 billion drives to equal one yottabyte of storage. Even if the government got the drives for $1 a drive, that's still $250 billion to be spending. On drives alone. That's not even taking into consideration any of the other costs associated with running a data center.

I think the news reports regarding the Utah facility are fabrication of real data presented by the US government to another media source and designed to invite the public into entertainment reagarding the power-house of funding. The article is not NPOV and should be edited by an expert from the Department of Justice. Fatum81 (talk) 02:58, 9 June 2013 (UTC)

After the expert from the Department of Justice edits this article for the government view, he or she should be called before Congress to testify under oath. ;-) Charles Edwin Shipp (talk) 13:19, 11 June 2013 (UTC)

Yes, Yottabytes!

"A yottabyte is so big as to be nearly unimaginable by casual computer users: It’s enough information to fill 200 trillion DVDs." . . .

"The companies participating in PRISM produce enormous amounts of data every day, so storing it would require computing power the likes of which the public has never seen. People who study technology and security believe that’s why the NSA [built] a million-square-foot data center near Salt Lake City."

http://www.theblaze.com/stories/2013/06/07/11-questions-you-probably-have-about-u-s-domestic-spying-answered/
Charles Edwin Shipp (talk) 14:09, 8 June 2013 (UTC)

Glad to hear that yottabytes was not a derogatory comment about somebody named Yotta.--Tomwsulcer (talk) 15:17, 8 June 2013 (UTC)

Insightful you are. :-) Charles Edwin Shipp (talk) 15:33, 8 June 2013 (UTC)

Gotta keep Wikipedia clean and wholesome and fun. :) --Tomwsulcer (talk) 15:49, 8 June 2013 (UTC)

Here are the prefixes you can use, from Binary_prefix:

• kb = 1 followed by 3 zeros (kilobyte, kb) — [‘k’ is lowercase]
• Mb = 1 followed by 6 zeros (megabyte, Mb)
• Gb = 1 followed by 9 zeros (gigabyte, Gb)
• Tb = 1 followed by 12 zeros (terabyte, Tb)
• Pb = 1 followed by 15 zeros (petabyte, Pb)
• Eb = 1 followed by 18 zeros (exabyte, Eb)
• Zb = 1 followed by 21 zeros (zettabyte, Zb)
• Yb = 1 followed by 24 zeros (yottabyte, Yb)

Yottabyte is the last one listed, so is the expected data even larger?
I’ll leave it to others to say how soon the NSA Utah Data Center will fill up.

Another measure is the Googol (100 zeros) — Charles Edwin Shipp (talk) 16:01, 8 June 2013 (UTC)

The facility is not designed to store a yottabyte of data. The cited Wired article by James Bamford does not even claim that. All it says is that "the Pentagon is attempting to expand its worldwide communications network, known as the Global Information Grid, to handle yottabytes (10²⁴ bytes) of data. That is the amount of data worldwide, and it is an apparent goal, directed to some unspecified date in the future, which could be on the order of decades. There is no possible way that the current facility could house a yottabyte with today's technology. COGDEN 07:48, 11 June 2013 (UTC)

Utah Data Center may become the largest, but it is not the only one: "The Utah Data Center is part of “a network of data farms” in the U.S. (Rick Bowmer/AP)"[1] — Charles Edwin Shipp (talk) 13:26, 11 June 2013 (UTC)

To avoid saying 'yottabytes' one computer scientist at the University of Utah says instead, "thousands of zettabytes". [2] Funny? Charles Edwin Shipp (talk)

Just in case you want to discuss if this dimension of data makes any sense: skeptics.stackexchange --MartinThoma (talk) 05:14, 3 July 2013 (UTC)

Thanks, that was interesting reading (skimming) and one 'take-away quote' is: "An NSA spokeswoman says the actual data capacity of the center is classified." Another two thoughts I had while reading/skimming: (1) to think about the size/capacity, ask Edward Snowden who has inside knowledge and indications are that such sites (multiple sites) are collecting every eMail etc and storing the text not just the metadata. (2) So how much would that take? (3) It is a military strategy to not reveal your power/capability. — Charles Edwin Shipp (talk) 13:35, 5 July 2013 (UTC) PS: The storage estimates for Facebook and particle accelerators were very interesting also.

Primary sources

Sources such as the NSA's mission statement, or remarks by generals, are essentially primary sources which can be used on some occasions but only with caution, and in this case, I think we need more references. A recent change says essentially that the NSA's own declaration of its own mission is factually correct. Here is the current wording:--Tomwsulcer (talk) 13:55, 18 April 2013 (UTC)

However, the primary mission of the National Security Agency is not domestic espionage, but rather protecting sensitive or classified US information from adversaries, sometimes called Information Assurance (IA), as well as collecting, processing, and disseminating intelligence information from foreign signals for intelligence and counterintelligence purposes and to support military operations, or Signals Intelligence (SIGINT).

My problem is we really need objective, reliable, secondary sources to back this up, and that we should not take the NSA at face-value here. Or, saying the NSA's declaration is, indeed, the case, is us committing original research, which is why I think we need to state that the NSA conceives its mission as such and such, but leave open alternate possibilities.--Tomwsulcer (talk) 13:55, 18 April 2013 (UTC)

Do we use this standard for any other government agency? That is, do we say that the Department of Agriculture "claims" its mission is such and such? Or do we say that is what it is? NSA's mission is codified in law, irrespective of what anyone else claims or thinks it does or doesn't do — I'm not sure how much more straightforward it gets. NSA's authorities originate from Executive Order 12333, and I have updated the article as such. das (talk) 03:12, 26 April 2013 (UTC)

Yes what makes Wikipedia believable and powerful is distance from the subject, impartiality. My problem with previous wordings (eg "Utah Data Center's mission is such and such, etc) is that it takes the point of view of a government source, as if our Wikipedia article on the Utah Data Center was essentially the same as the government website's version. It is more professional for us to say, as Wikipedia contributors, that 'Point X was made by Source Y' rather than 'Point X is true'. Can you see why? Saying 'Point X is true' weakens our authority here, in my view. See, we're not a government source (or at least I am not); let the reader make up their mind about what to believe. And my personal sense is that it would be difficult for any government agency to strictly separate foreign intelligence from domestic data, that these will get mixed in together regardless of its supposed mission, so regardless of what the NSA says its mission is, I think that such data will be in there. My personal POV is that this is a good thing, that this is something the US should be doing provided that privacy protections are built in. Like, such data could have helped US intelligence agencies prevent the Boston Marathon bombing.--Tomwsulcer (talk) 10:42, 26 April 2013 (UTC)

The first sentence of the National Security Agency article on Wikipedia says:

The National Security Agency (NSA) is a cryptologic intelligence agency of the United States Department of Defense responsible for the collection and analysis of foreign communications and foreign signals intelligence, as well as protecting U.S. government communications and information systems

There is no "government website version" of the Utah Data Center. Frankly, the paragraph in "Possible Purpose" that you earlier questioned being added to this page has nothing to do with the Utah Data Center. Impartiality and NPOV is one thing, but NSA's mission is what it is. If people want to claim or think it's doing something else, that is fine, but it's not a matter of "according to the National Security Agency...", it's according to US statute. The mission is a factual thing, and is not POV. We can get into all sorts of semantic discussions, or ask what the internal security apparatus in totalitarian states says their "missions" are as if it's on the same footing as claims made by government in an open society, and as a result lose sight of the forest for the trees. The fact of the matter is that in the public domain, no one -- including whistleblowers who weren't even at NSA when the Utah Data Center project was conceived -- has any idea what is happening at the Utah Data Center beyond public statements by government, which is that this is an Intelligence Community (not NSA) facility to support the Comprehensive National Cybersecurity Initiative, for which NSA is the executive agent. This is present in contract award and other documents referenced in some of the articles linked here. Executive agents are often used in DOD and the IC because someone with expertise has to be responsible for the project, but this is an IC resource, not an NSA resource -- there are tips to this all over, such as the fact that the Director of National Intelligence seal is present on the sign at the entrance to the facility. As for the whole issue of how to target only foreign communications in a sea of domestic data, or how to target non-communications traffic for e.g. cybersecurity and intelligence purposes, there is very little discussion other than from government sources, because the vast majority of the information is classified to protect intelligence sources, methods, and capabilities. General Hayden's speech at the national press club and the NSA General Counsel's recent speech at Georgetown University are absolutely the best and most detailed and complete unclassified discussions of these issues (if you have a personal interest, please read both of these). The point is that on contentious issues where the facts are classified, it is hard to present the balanced view you seek to find, and these are all issues of protracted discussion which don't really have a place on this page. As it stands, this article is actually pretty terrible in terms of representing an unbiased view based on the facts that are out there -- when you asked whether some portions of the initial "Possible Purpose" paragraph that didn't relate to the Utah Data Center should be removed, the answer would be that it all would have to be removed, because none of it has anything to do with the Utah Data Center, but rather allegations relating to NSA as a whole. Any links of such allegations to the Utah Data Center are gaps filled in by the imagination or assumptions of people with no direct knowledge. How is that any more reliable or unbiased? das (talk) 13:04, 26 April 2013 (UTC)

Perhaps, but I'm unconvinced. What is the harm of adding the phrase according to the National Security Agency to the line about the mission? I read over the NYT article and you are right that the Utah Data Center is not explicitly mentioned, but I think it is still a valid source since the NSA runs the Utah Data Center, and the data stored in it is likely to contain what the whistleblower was referring to. I agree there is much rumor and speculation about what, precisely, the Utah Data Center is supposed to do, or what its limits are; does this mean we have to implicitly trust the government's supposed statement of its purpose? I'm skeptical. I grew up during the Vietnam War with presidents lying about an entire war. What I am saying is that it makes our contributions stronger, and more believable, if we would put more distance in it, not saying the Utah Data Center's purpose is such and such, but rather that the NSA says its purpose is such and such; reporters reading this article will have more respect for what we do here if we phrase it like I suggest. Further, what is weird is that a data facility like the one being built in Utah is something similar to what I recommended being built, in an essay I wrote years back called Common Sense II, which advocated such a facility to permit domestic surveillance (but with explicit approval from citizens, along with privacy protections, as a way to prevent terrorism) as a way to help investigators notice possible dangers to the public, and to protect people. What I am saying is there is probably a strong incentive for government to use this facility for such a purpose, particularly after the bombings at the Boston Marathon. So, my POV, which of course must be kept out of the article, is that this facility is being used (or will be used, or should be used) to collect domestic information. FYI.--Tomwsulcer (talk) 17:35, 26 April 2013 (UTC)

I think my point is that it is accurate to say that NSA's mission is such and such (because it is, irrespective of what it may or may not be doing), and fine to say that NSA says the Utah Data Center is for such and such. It is also accurate to say it is against the law to target, collect, store, analyze, or disseminate the content of the communications of a US Person without an individualized warrant. That means that if the Utah Data Center ends up collecting on US Persons, it will be doing so in direct violation of the law. But there is a lot of complexity here, because in order to examine traffic for threats (whether it is malicious attacks against US assets or foreign communications constituting a threat), obviously the capability to examine said traffic must be present. Again, as General Hayden noted:

Gone were the days when signals of interest -- that's what NSA calls the things they want to copy -- gone were the days when signals of interest went along some dedicated microwave link between strategic rocket forces headquarters in Moscow and some ICBM in western Siberia. By the late '90s, what NSA calls targeted communications -- things like al Qaeda communications -- coexisted out there in a great global web with your phone calls and my e-mails. NSA needed the power to pick out the one, and the discipline to leave the others alone. So, this question of security and liberty wasn't a new one for us in September of 2001. We've always had this question: How do we balance the legitimate need for foreign intelligence with our responsibility to protect individual privacy rights?

So it's fine for people to think, "Well, I still believe that the Utah Data Center is going to be used to illegally spy on US citizens"...but there is no proof of that, and it would also be patently against the law. Even Binney -- who, again, hasn't been at NSA for over a decade -- and Drake in recent interviews don't say the Utah Data Center "will" be used for monitoring Americans; just that building such a storage capability means it could be used for that if the "wrong person" ever came into power and decided to abuse it. Okay, great...but any government power can be abused. Do we talk about how the US military "could" be turned against US citizens, as if it is a foregone conclusion? This is also an argument against technology, when it is the rule of law, not the technology, that is paramount.

As for the sentence in question, I had changed it to read, "[...] Executive Order 12333, which governs United States intelligence activities, defines NSA's missions as [...]," which is a factual, NPOV statement. das (talk) 01:31, 27 April 2013 (UTC)

Fair enough. Seems like we have spent enough time dickering over this small point.--Tomwsulcer (talk) 11:59, 27 April 2013 (UTC)

The NSA was awarded a blanket warrant by the FISA court to spy on all american citizens. When they say "we aren't illegally collecting every citizen's data" they mean "it isn't illegal because we have a warrant", not "we aren't doing it". The tone of this article implies the biased insincere frame of the NSA, when the objective frame is not in contention. 2603:7081:4F07:3962:C4AE:7153:E305:D15A (talk) 13:27, 19 April 2023 (UTC)

News references

If useful, please use the following for references...

• Government is tracking all U.S. phone calls; LA Times; June 6, 2013. • Sbmeirow • Talk • 03:51, 7 June 2013 (UTC)

• How Congress unknowingly legalized PRISM in 2007 Washington Post.--Tomwsulcer (talk) 13:04, 7 June 2013 (UTC)

Codename Bumblehive [light-parody, not a government site]

The state flag of Utah highlights a beehive, — motto means industriousness. The website (apparently from the US government) http://nsa.gov1.info/utah-data-center/ identifies the codename and mentions 'Bumblehive' 24 times. FYI, Charles Edwin Shipp (talk) 01:37, 9 June 2013 (UTC)

It's a .info site, not a .gov site, so it's likely not a US govt. site. And a disclaimer at the bottom of the page reads: "This is a parody of nsa.gov and has not been approved, endorsed, or authorized by the National Security Agency or by any other U.S. Government agency. Much of this content was derived from news media, privacy groups, and government websites. Links to these sites are posted on the left-sidebars of each page." 01:57, 9 June 2013 (UTC)

Thanks. I wondered. I've added [not a government site] to the 'External links' section/line. They have excellent pictures, info, and Salt Lake Tribune articles. I wanted to know why the following message appeared when you go to the site, and now we know: [paraphrasing: by coming to this site . . .] has been removed. Instead, we read at the bottom "This is a parody of nsa.gov and has not been approved, endorsed, or authorized by the National Security Agency or by any other U.S. Government agency.

Much of this content was derived from news media, privacy groups, and government websites. Links to these sites are posted on the left-sidebars of each page." So I would conclude that 'Bumblehive' was never a military codename. Charles Edwin Shipp (talk) 14:09, 9 June 2013 (UTC)

Original Research

I removed a bunch of original research (mostly stuff about FISA and the law, which isn't really about this article. Basically the entire "possible purposes" section was OR.) However, one sentence I removed seems like it should still be included. But the source it was cited to was incorrect, so I was wondering if anyone knows the real source so we can add the sentence back in. The relevant text is: NSA whistleblower William Binney alleged that the Bluffdale facility was designed to store a broad range of domestic communications for data mining without warrants. Capscap (talk) 08:16, 9 June 2013 (UTC)

Here is a reference implying use: "NSA Whistleblower Speaks Out on Verizon, PRISM, and the Utah Data Center" [3] — Charles Edwin Shipp (talk) 14:12, 9 June 2013 (UTC)

Actually, I should correct myself—although this website seems to have good pictures, good info, and good links to standard newspapers (like the Salt Lake Tribune, newspaper) it is a satire website. (LibertasUtah, {only in Utah}, coined the nickname "Bumblehive" for the beehive state NSA Utah Data Center.) Be wary of falling into fabrication. — Charles Edwin Shipp (talk) 00:40, 24 June 2013 (UTC)

Data Center in the news

There might be some new information here:

• http://fox13now.com/2013/06/06/nsa-data-collection-turns-attention-to-utah-data-center/
• http://www.nytimes.com/2013/06/09/us/revelations-give-look-at-spy-agencys-wider-reach.html?pagewanted=all

Capscap (talk) 08:24, 9 June 2013 (UTC)

If you read the following article,

• http://www.theblaze.com/stories/2013/07/01/seven-stats-to-know-about-nsas-utah-data-center-as-it-nears-completion

a reader makes this correction: "For the record, they are Intel “Xeon” processors. Not “Xenon Core” processors."

Charles Edwin Shipp (talk) 14:09, 2 July 2013 (UTC)

Other similar data centers, some forming a network

Here's an interesting item showing that Apple Co. needs support for its iCloud and iPhone/iPad.
http://blogs.wsj.com/digits/2013/07/01/apple-invests-in-solar-farm-for-nevada-data-center/?mod=trending_now_5
Charles Edwin Shipp (talk) 19:09, 2 July 2013 (UTC)

Our reference (to the Thomas Burr piece) at the bottom of our WP article notes that, "The Utah Data Center will be part of NSA’s interconnected network that includes sites in Colorado, Georgia and Maryland, and since the Utah facility will be the largest, there is a good chance Americans’ phone call data could land in the Bluffdale site at least temporarily. "I wouldn’t say I know it for a fact," says Steven Aftergood, director of the Federation of American Scientist’s Project on Government Secrecy. But "when you build a facility of that scale, it’s probably meant to be used, and the storage and processing of large volumes of collected data would seem to be a plausible use of this facility." — Charles Edwin Shipp (talk) 19:19, 2 July 2013 (UTC)

Lynnette and I will be traveling from Utah Valley to Salt Lake Valley via 'Point of the Mountain' pass. I plan to stop and see what I can find out about the openhouse and if there will be tours. We were at the Pentagon and there were tours, but you needed reservations with a three-day advance notice to for clearance checking. Are there other questions I should ask the guard or the voice system? Charles Edwin Shipp (talk) 03:07, 7 August 2013 (UTC)

This just in . . .

Because of Edward Snowden leaks, expect a lot of new information to come in between now and the NSA Utah Data Center opening (said to be in October, but some articles say "September or October": probably operational in September with ribbon-cutting & public tours in October.) I'll be there since my wife and I have a lot of family in American Fork, Utah County, and Salt Lake County. FYI, the facility is on the county line; on Camp Williams Army Base, next to the state prison.

This just in:

• "Experts worry about foreign, domestic implications of Utah Data Center" [4] July 2, 2013, Salt Lake Tribune.

Charles Edwin Shipp (talk) 19:40, 2 July 2013 (UTC)

Article title

• This is text from the Camp Williams article that was deleted (I have restored) but has several older refs re the 'Utah Data Centre', which was apparently previously called Community Comprehensive National Cybersecurity Initiative Data Center . [anon]

• I think this article should be called 'NSA Utah Data Center' since Utah has many data centers. Charles Edwin Shipp (talk) 15:58, 8 June 2013 (UTC)

• Call it the 'ICCNCI Data Center' from the formal name listed at the top or our article: Intelligence Community Comprehensive National Cybersecurity Initiative Data Center". -- Charles Edwin Shipp (talk) 15:51, 7 August 2013 (UTC)

I moved these comments orphaned at the top without a subhead.96.227.66.159 (talk) 23:36, 10 July 2013 (UTC)

Water Usage

A local story from Bluffdale. The center is expected to use 1.7 million gallons of water per day when fully operational.

http://www.ksl.com/?sid=25978926&nid=148

--71.20.55.6 (talk) 16:14, 13 July 2013 (UTC)

Projected Advances in HDD Capacity to 60TB in 2016

For the Great Yottabyte debate.

http://hothardware.com/News/Hard-Drive-Capacity-Could-Increase-to-60TB-by-2016-IHS-iSuppli-Says/ --71.20.55.6 (talk) 07:27, 14 July 2013 (UTC)

Calculating Storage Capacity from Power Usage

This may or may not qualify as Original research. Nonetheless, I'm including it here.

The center is expected to use 65 Megawatts of electricity. By knowing the power consumption of large capacity drives, you can put an upper limit on the storage by assuming that all electricity used will power HDD drives. Much of the power will go to cooling, and equipment besides drives, of course.

The lowest average operating power consumption for 4TB Seagate Enterprise drives is 11.27Watts. This results in a storage capacity of 23 Exabytes.

Assuming that capacity is more important than speed, a lower power model using only 6.49 Watts can be found resulting in 40Exabytes. Assuming that 60TB drives show up a few years down the line that only use 6.5 watts, the capacity would be around 600Exabytes.

Switching to SSDs is a costly proposition, but significantly reduces the power consumption per byte (and also the actual dimensions are less, and the drives are a much faster). A 960GB model that uses 0.6 Watts would result in 104Exabytes.

Whatever the scenario, the storage capacity of that facility should still expected to be in Exabytes, rather than Yottabytes, given available technology.

--71.20.55.6 (talk) 20:47, 15 July 2013 (UTC)

Power consumption is probably not a useful measure of storage capacity. If they use hard drives they would only need to have a few of them turned on to read and write while the rest are not consuming power. They could conceivably be using modern magnetic tape storage, similar to the way the LHC's experiments store their data. Pulu (talk) 22:13, 21 July 2013 (UTC)

References

How about "thousands of zettabytes" or millions of exebytes? — Charles Edwin Shipp (talk) 00:20, 16 July 2013 (UTC)

It has to be under a Zettabyte. Neither magnetic platters nor flash can accommodate Yottabytes if that power figure is right. The limit on Flash is around 16TB, but that's not till the end of the decade. Even then we still get low-order Zettabytes. Yottabyte might be the whole size of the whole Intelligence operation. But not all in Bluffdale.

http://cseweb.ucsd.edu/users/swanson/papers/FAST2012BleakFlash.pdf

--71.20.55.6 (talk) 01:52, 16 July 2013 (UTC)

More than that, half of the power goes to cooling, another 11% or so to UPS loss, 3% to lighting leaving only 36% Everything else. it cuts us down to "tens of Exabytes" at most. For the next several years.

https://www.youtube.com/watch?v=w3kuhVktCOE

--71.20.55.6 (talk) 20:33, 16 July 2013 (UTC)

That said, William Binney believes the capacity may be on the order of 5 Zetabytes. He's made this claim at least twice, from what I've seen in videos of his speeches and interviews. Most recently 7-19-2013. http://english.cntv.cn/program/newsupdate/20130719/104150.shtml

--71.20.55.6 (talk) 09:39, 21 July 2013 (UTC)

Explaining the Binney estimate: Binney based his estimate on a misread of marketing material he interpreted to mean that a 10 exabyte portable datacenter would require 21 racks and take up 200 square feet. He was mistaken in that interpretation. --71.20.55.6 (talk) 07:21, 29 July 2013 (UTC)

The Diagram

The diagram of the data center for this page roughly matches the diagram from wired magazine. But neither of these resemble the photos of the facility seen here and here. There is only one row of buildings and the data halls are oriented differently. These two photos are consistent with the photo on the wiki. Pulu (talk) 22:26, 21 July 2013 (UTC)

Forbes has obtained blueprints 7/24/13

Brewster Kahle estimate: 12 exabytes. Paul Vixie estimate: 3 exabytes

http://www.forbes.com/sites/kashmirhill/2013/07/24/blueprints-of-nsa-data-center-in-utah-suggest-its-storage-capacity-is-less-impressive-than-thought/

--71.20.55.6 (talk) 02:31, 25 July 2013 (UTC)

Forbes story- Comments section - OK to upload Blueprints to wikipedia

Anon 3 days ago

How does one go about obtaining permission to upload the blueprints to Wikipedia for the article about the Utah datacenter?

   Called-out comment

Reply Author Kashmir Hill Kashmir Hill, Forbes Staff 3 days ago

They are government docs so I don’t think we’d claim ownership of them.

   Called-out comment

http://www.forbes.com/sites/kashmirhill/2013/07/24/blueprints-of-nsa-data-center-in-utah-suggest-its-storage-capacity-is-less-impressive-than-thought/ — Preceding unsigned comment added by 71.20.55.6 (talk) 07:16, 29 July 2013 (UTC)

William Binney's role in Stellar wind

In August 2012, The New York Times published short documentaries by independent filmmakers entitled The Program,[7] based on interviews with a whistleblower named William Binney, a designer of the NSA's Stellar Wind project.

This is a bit misleading. Binney designed a program called Thinthread, the backend of which was later bastardized by a third party contractor and integrated into Trailblazer, and again bastardized and inserted into Stellar Wind. Binney had left NSA by the time Stellar Wind got going. It's not entirely unfair to categorize Binney as a designer of Stellar Wind, (and he sometimes describes himself as such) but it's not a complete picture. Stellar Wind was the reason for the James Comey rebellion in 2004, in which Ashcroft, gravely ill in hospital, made the decision to allow Comey to kill the project. The primary sources are Binney's Interviews, one with Laura Poitras, referenced above. Tom Drake is another source.

There is a question of how dead Stellar Wind really is:

http://www.washingtonpost.com/investigations/us-surveillance-architecture-includes-collection-of-revealing-internet-phone-metadata/2013/06/15/e9bf004a-d511-11e2-b05f-3ea3f0e7bb5a_story.html --71.20.55.6 (talk) 19:38, 29 July 2013 (UTC)

In my opinion it isn't correct to say that mr. Binney is a designer of Stellarwind. Initially it was thought that Stellarwind was a program or a system for intercepting communications, but the report of the NSA inspector general from 2009, which was disclosed by Edward Snowden last June, showed that Stellarwind is actually the codename of a so-called security compartment for information related to the President's Surveillance Program (PSP). As far as Thinthread and Trailblazer were also part of the PSP, information gathered through those programs may have been classified under the Stellarwind codeword. P2Peter (talk) 02:24, 21 October 2013 (UTC)

Sparks fly?

If you do a Google-search with the search terms, [ Utah data center openhouse october 2013 ] you will see that the opening ceremony for the Utah Data Center [NSA Utah Data Center] has been postponed, if you can believe the information found by Wall Street Journal (WSJ), Reuters, and the Salt Lake Tribune. In my skimming yesterday, I saw the the group protecting the constitution "Restore the Fourth [Amendment], Utah"[5] marched in protest a few days ago, (can you say sparks were flying?) and a day or two ago, it was announced that the October ribbon-cutting ceremony would be postponed for maybe a year, if you believe it. [6]

In a Salt Lake Tribune article, one reporter/writer notes that the Utah Data Center may be open already. [7]

If you do the Google-search suggested, a favorite site of mine has excellent leads but is a parody of the NSA.gov site: (so view it for leads, but not for referencing here): =[;-)  :: [8] They have excellent pictures of the site! Etc .!.

From NSA, their bottom line: "Washington • The National Security Agency says electrical problems at its Utah Data Center were not as dire as reported earlier this week, didn’t damage any expensive computer equipment and shouldn’t delay the opening of the massive storage site. Our current assessment is this issue will be fully resolved, mission systems will be installed on schedule, and the project will remain within budget," NSA’s director for installations and logistics, Harvey Davis, says in a letter sent to congressional intelligence committees this week. The Salt Lake Tribune obtained a summary and discussion of the letter." [9] -- Charles Edwin Shipp (talk) 09:11, 14 October 2013 (UTC)

I wish I'd seen you post earlier as I have just added a section referring to the 'alleged'(?) electrical problems.

A senator from Utah wants to make it illegal to supply water to the facility! Clever dick! See Utah Republican To Cut NSA Facility’s Water Supply 220 of ^Borg 09:22, 9 March 2014 (UTC)

datacenter image

Hi, should we replace older map with more accurate verison? `a5b (talk) 02:42, 31 December 2013 (UTC)

older

new version

I don't actually know how the center really looks like, but the newer version does look neater and more visually appealing to me. — Preceding unsigned comment added by A1candidate (talk • contribs)

There are blueprints published by Forbes - http://www.forbes.com/sites/kashmirhill/2013/07/24/blueprints-of-nsa-data-center-in-utah-suggest-its-storage-capacity-is-less-impressive-than-thought/ and there are photos of center: http://www.dailymail.co.uk/news/article-2337420/Utah-Data-Center-The-million-square-foot-Utah-data-mining-facility-built-NSA.html (warning, the Conceptual Site Plan is not accurate for buildings positions) `a5b (talk) 13:45, 31 December 2013 (UTC)

Yes, I've been there. --  Done -- Looks good. -- TNKS, AstroU (talk) 23:08, 7 February 2015 (UTC)

Defending America in cyberwars

The defenders against cyber-attacks draws other attacks to the Utah gov't even though not attached.

Headline-1: Massive Utah cyberattacks — up to 300 million per day — may be aimed at NSA facility

• http://www.sltrib.com/news/2135491-155/massive-utah-cyber-attacks-may-be

QUOTE: "10,000-fold increase » Since the facility was built, Utah government has had up to 300 million attempted attacks a day." -- AstroU (talk) 23:19, 7 February 2015 (UTC) -- PS: FYI for future editing. New NEWS today, for future editing

Concerns and criticism of mass-surveillance continues

Concerns, criticism, and also ambivalence, of mass-surveillance continues with citizens in Utah and across the nation, according to this subscription WSJ article.
Headline: A Top-Secret NSA Site Draws Swipes, Shrugs

• http://www.wsj.com/articles/a-top-secret-nsa-site-draws-swipes-shrugs-1430523735

QUOTE: "Utah’s reaction to data center mirrors wider ambivalence over surveillance programs -- As a defiant statement against what it sees as government overreach, a group of Utahns “adopted” the desert highway that leads to the National Security Agency’s secretive and sprawling new facility in Bluffdale." -- Wall Street Journal (online) give this brief thumbnail, but requires a subscription to read the entire article. -- Narnia.Gate7 (talk) 00:45, 2 May 2015 (UTC)

...is expected to...

As of October 2019, why are there so many phrases that start with "is expected to," and why is the facility referred to as "projected," if it's already been built? 76.189.141.37 (talk) 23:59, 5 October 2019 (UTC)

Look at the dates of the references: 2011, 2012, etc. That was the information then. So the article needs updating; are you up to the task?--Tomwsulcer (talk) 01:28, 7 October 2019 (UTC)