Talk:Terabyte/Archive 2

Terminology
It would help computer users like me, if we handled the difference between GB as reported by system software like Windows (i.e., 2^30 = 1,073,741,824) and GB as reported by many device manufacturens (i.e., 10 ^ 9 = 1,000,000,000).

Unfamiliar terms like GiB and TiB do not help much. Being too precise loses much of our audience. --Uncle Ed (talk) 18:20, 20 December 2007 (UTC)


 * The point is discussed in Gigabyte. Is that what you mean? Thunderbird2 (talk) 18:52, 20 December 2007 (UTC)

Yes, exactly! In fact, I'd also like to have a table which compares Giga BITS vs. Giga BYTES. When uploading and downloading large files, people want to estimate how long it will take.

For example, if my AVI video file is 12 GB (gigabytes) in size, how long will it take to transmit on a 100Mb (megabits per second) computer network? --Uncle Ed (talk) 19:14, 20 December 2007 (UTC)


 * It's fairly safe to assume that a megabit as one million bits, but a gigabyte is anyone's guess. I suppose that's your point, right?  How do you think the article can be improved though? Thunderbird2 (talk) 22:30, 20 December 2007 (UTC)

For one thing, remove claims that tebibyte et al. is in wide use. *This amount is now known instead as a tebibyte, to avoid confusion.

Secondly, let's collect the various charts such as the one at Binary prefix which compare byte prefixes based on 2^10 (1,024) with those based on 10^3 (1,000). I'm getting tired of re-inventing the wheel here. --Uncle Ed (talk) 02:38, 21 December 2007 (UTC)


 * Where do you see the words "This amount is now known instead as a tebibyte, to avoid confusion"? —Preceding unsigned comment added by Thunderbird2 (talk • contribs) 08:06, 21 December 2007 (UTC)
 * Oh, I see now - you had already removed them. I have reinstated a weaker version of the claim. Thunderbird2 (talk) 08:16, 21 December 2007 (UTC)

Thanks, T-bird. Your wording is much better than my deletion. Merry Xmas! --Uncle Ed (talk) 01:32, 22 December 2007 (UTC)

Merry Christmas to you too :-) By the way, have you seen this? Thunderbird2 (talk) 18:52, 25 December 2007 (UTC)


 * I made some further edits to indicate that both usages (SI and binary) are still in use; someone had edited it as if SI was the last word and everyone had ceased using the binary definition. (Yes, that would be nice, and it would be nice if the "tebi" style prefixes would come into use, but it's not the reality.  I also addded that operating systems typically report the binary version.  58.108.76.51 (talk) 01:08, 23 April 2008 (UTC)

Mac Pro comment
I'm removing it, as it doesn't need to be in the article. Any computer can have 4TB worth of 4 harddrives. 68.154.166.169 (talk) 14:34, 14 April 2008 (UTC)
 * I agree. 200.192.77.252 (talk) 20:56, 16 April 2008 (UTC)

"Terabytes in use"?
I not sure the relevance of the "Terabytes in use" section. Terabyte hard drives are widely commercially available and many PC users (myself included) have such drives. The section may have been important a few years ago, but is now outdated and needs to be removed.--Marcus Brute (talk) 04:32, 18 January 2009 (UTC)
 * I partially agree, and have removed some of the old references. But not entirely. The point of the section is not to say "look, terabytes are HUGE!" but to give an estimate of how much a terabyte is, in "real-world" terms. For this purpose, some of them are relevant, I think. Also, since you placed the tags at the top of the page — can you describe what the issues are? Shreevatsa (talk) 15:20, 18 January 2009 (UTC)


 * The problem here is that other editors are likely to dump an endless list of examples as there is no selection criteria (see WP:TRIVIA). This could be solved by re-writing the list into prose or adding an initial criteria for the list such as "Examples of the term terabyte being used in different fields are:", consequently examples would be encouraged for computer hardware, printed information, video compression... with only one example for any possible field. In particular this eliminates the endless list of computer hardware examples. I have implemented the latter solution without removing any examples in the current list.—Ash (talk) 06:04, 31 July 2009 (UTC)

Putting the size in perspective
Using the binary counting numbers for megabyte and terabyte, it would take 209,714.8 five megabyte hard drives to hold one terabyte. Did that many 5 meg drives ever exist? —Preceding unsigned comment added by Bizzybody (talk • contribs) 05:26, 27 August 2009 (UTC)

yes, the IBM PC's at one point, had 5 meg hard disks, I'm sure they shipped millions of them. OldCodger2 (talk) 20:09, 12 December 2012 (UTC)

Access dates in references
Hi, The edits you reverted are specifically covered by MOS:DATE, you are adding ambiguous dates back onto the page. Please take this up on the article talk page if you feel this is wrong.—Ash (talk) 00:21, 15 November 2009 (UTC)
 * The dates are NOT ambiguous, these are the standard ISO formats, see ISO 8601, never ambiguous. There is no other format of this type that uses another sequencing. WP doesn't use them in running text, because it is not common in English to do so, but this doesn't mean they shouldn't be used for technically accurate representations elsewhere. Kbrose (talk) 00:26, 15 November 2009 (UTC)

MOS:DATE does not prohibit them, in fact it states they are useful for technical reasons. Kbrose (talk) 00:33, 15 November 2009 (UTC)
 * MOS:DATE represents the consensus and is very clear. You are free to propose a change so it allows for ambiguous dates in the style you prefer. The argument you have given above would also allow for the use of the "YYYYMMDD" format which ISO 8601 uses but would hardly be meaningful in footnote dates in encyclopaedic articles (Wikipedia is an encyclopaedia, not a technical document). I'm off to bed but if you still feel the same way in 12 or 24 hours, I suggest a third opinion may be in order. Oh, I can see you just reverted the text again. Isn't that rather rude whilst we are still discussing the matter?—Ash (talk) 00:54, 15 November 2009 (UTC)
 * The ISO format is not ambiguous and a strong argument can be made for its use in this article due to its popularity in technical/scientific fields. --Cyber cobra (talk) 01:01, 15 November 2009 (UTC)


 * There is absolute no consensus whatsoever in MOS to wholesale-replace the format where used for good reason. Cataloging of access dates and other similar technical uses and comparisons are certainly good reasons. Kbrose (talk) 01:12, 15 November 2009 (UTC)


 * There is consensus to apply the table in MOS:DATE to reduce ambiguity of dates. The revert in question was limited to the date and accessdate parameters used in the cite news and cite web templates (see diff). In the case of the cite news template it states "date: Date of publication. To avoid ambiguity, write out the month in words, using the same date format as in the main text of the article" and in the case of the cite web template the many examples used in the template guidance use the unambiguous date format in compliance with MOS:DATE. Consequently I am only trying to apply the templates as defined by the guidance for the templates and the Wikipedia manual of style. If you consider the policy of consistency, the footnotes should align with the date format in the sources or the body of the text; you can see that text months are used rather than numeric months, this would also be a rationale for the footnote to use the textual month.—Ash (talk) 01:22, 15 November 2009 (UTC)
 * The policy says nothing about body dates being consistent with citation dates, only that each type itself should be consistent. --Cyber cobra (talk) 01:28, 15 November 2009 (UTC)
 * I also point out the failure of Mosnum/proposal on YYYY-MM-DD numerical dates. --Cyber cobra (talk) 01:28, 15 November 2009 (UTC)
 * Er, please read the text in italics in my above paragraph, the template guidance states the exact opposite.—Ash (talk) 01:33, 15 November 2009 (UTC)
 * That's one template vs. the several in use; examples in docs aren't binding. The guidelines conflict, an unfortunately not uncommon occurrence. --Cyber cobra (talk) 01:49, 15 November 2009 (UTC)


 * I note that this section title refers to accessdate rather than date. To avoid conflating the two, if we put aside the debate on accessdate format, as the guidance and templates for citations clearly state that citation date should use the same format as the article body, is there any argument against applying policy here? I note the conclusion of the Mosnum proposal you referred to was "consensus to allow YYYY-MM-DD in footnotes" (not to mandate it) but the context of the discussion itself principally related to accessdate and there was no consensus to obviate the current policy of having date in the same format as used in the article.—Ash (talk) 12:15, 15 November 2009 (UTC)

What now, 1000 or 1024?!?
The article states that "1 terabyte is 1000000000000 bytes or 1024 gigabytes". Both of those obviously can't be correct (that would imply that a GB is 976 562 500 bytes, which is nonsense both from a SI and from a binary viewpoint). Time to decide?

I'm changing 1024 to 1000. — Preceding unsigned comment added by 212.85.182.240 (talk) 07:27, 12 May 2012 (UTC)

High ISO IEEE number -- makes no sense
Huh? could somebody please explain the following quote, it makes no sense to me. The High ISO IEEE number of the terabyte means it can be thought of a physical form of memory, of which the singular byte cannot be.  I nominate this for deletion or rewrite, any takers? OldCodger2 (talk) 20:16, 12 December 2012 (UTC)

Cost of Storage
I added a new section on Cost of storage. I hesitated to do this without discussing it first, however I have often seen statements of encouragement for people to be somewhat fearless in making contributions. I feel that a section on cost is another way to add context to the understanding of the size of a terabyte. Since this measurement (terabytes) is most commonly used with reference to disk drives and since there is already a section on disk drive capacity, it seemed reasonable that there should also be a context to show how the cost of that capacity has changed over time. Does that make sense? Now, if I could just remember how much I used to pay for a box of punched cards... ;-) OldCodger2 (talk) 20:26, 12 December 2012 (UTC)

After seeing a 3 TB drive on a 3 day sale for $135 i was inspired to update the Costs table with 2013 prices. The problem is that this is a simplification, the prices are actually all over the place depending on features and speeds and there are older drives still being sold for more than newer drives etc.. I tried to pick drives that were representative of the over-all trends. In the case of the 1 TB drive I actually split the difference, there was a cheap drive for $65 but of questionable quality, and a better drive for $75, so I went with $70 as being a representative price. Having actually paid $1000 for a 1 Gigabyte drive... a great bargin at the time... these prices just continue to astonish me. OldCodger2 (talk) 02:51, 17 April 2013 (UTC)

Illustrative usage examples
I came here to see if I could find some layman's examples of what 1TB is equivalent to (x no. of books, say, or y hours of video). Only the 'Audio' example in the list here does that - the rest just link one un-visualisable quantity to another. — Preceding unsigned comment added by 85.210.132.93 (talk) 10:59, 15 August 2014 (UTC)

The information about there being 160 TB/s of Internet traffic in 2008 is wrong. The source provided does not mention this at all, while the graph in the source shows just above 10,000 Petabytes per month which is around 3 TB/s. Per Cisco, 2014 only had 16 TB/s of traffic. http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/VNI_Hyperconnectivity_WP.html I'd fix this, but I don't really know what to put as the numbers were chosen as a comparison. — Preceding unsigned comment added by 2601:484:C201:7431:21D7:1DBF:575B:967A (talk) 05:26, 9 September 2015 (UTC)

The JEDEC link
This JEDEC reference clearly shows "binary tera = 1099511627776 bytes" As such there isn't any logical justification for removing it from the article. Fnagaton 05:32, 11 July 2015 (UTC)


 * See the talk here Fnagaton 05:54, 11 July 2015 (UTC)


 * Dondervogel2 is vandalizing the article by repeatedly blanking an entire section. . Please stop this vandalizing otherwise I will have not alternative to raise the issue of your vandalism. Glider87 (talk) 11:54, 22 July 2015 (UTC)
 * On the contrary, it is you and your sockpuppet Fnagaton who are disturbing the relative peace and stability of the article that existed before your recent reemergence on the scene. You have been warned of your activities on your talk page User talk:Glider87. Your edits introduce your personal point of view and wish that the metric prefixes remain in use in the binary sense, despite standards based rule and the steady progress of acceptance of binary prefixes across the industry. Kbrose (talk) 13:44, 22 July 2015 (UTC)
 * I'm not a sockpuppet, I'm an editor who has seen the problematic point of view pushing of Dondergoval2 and your edits also push a similar same point of view. Vandalism of articles, which Dondervogel's edits are, and reverting of vandalism is not edit warring. You have no consensus for your edits because your edits violate WP:NPOV. Your edits violate WP:NPOV because you are trying to remove relevant sourced material from the article because it is contrary to your stated beliefs. You earlier said people using terabyte with binary quantities were lazy, which shows your point of view pushing. Glider87 (talk) 14:06, 22 July 2015 (UTC)
 * The original reference clearly does not show a binary "Tera-". I haven't checked any other references in this thread.  That being said, Dondergoval2 and Kbrose have damaged other articles because of their (apparent) belief that binary "giga-" should not be used.  — Arthur Rubin  (talk) 15:04, 22 July 2015 (UTC)
 * @Arthur Rubin: Thank goodness a real editor finally appears on this page. Thank you for confirming what Kbrose and I have repeated multiple times. To be honest though I don't get the second part of your message, concerning "damage" caused by "Dondergoval2 and Kbrose".  Are you able to substantiate this accusation? Dondervogel 2 (talk) 15:36, 22 July 2015 (UTC)


 * Actually the table in the standard document does show terabyte worth a binary power of two value. I think you Dondervogel and Kbrose should be blocked for edit warring and damaging the article.Fnagaton 05:46, 23 July 2015 (UTC)
 * User:Glider87 posted a screenshot from the JEDEC standard document JES100B it shows that the table includes a "tera + binary" which is equal to a power of two value. Including that table in the standards document is a reliable source. Fnagaton 15:11, 23 July 2015 (UTC)
 * Dondervogel 2 is edit warring and refusing to talk here. The reliable sources do support what is in the article and they are not dubious.Glider87 (talk) 13:26, 27 July 2015 (UTC)
 * The JEDEC link (link number 2) has the table "Factor Name Symbol Origin Derivation" that table contains the same information as the picture from the standards document. Trying to claim that link does not show binary tera is a huge mistake because that table and the picture are from the same source. The text in the table is badly formatted that's all, but it's still reliably showing binary tera. Glider87 (talk) 13:32, 27 July 2015 (UTC)
 * The other standards document has binary terabyte used. These links to the JEDEC all show binary terabyte being used. Calling them dubious and repeatedly putting dubious tags in the article is not-neutral. If you think the links are not good then you explain here why not. Don't damage the article with incorrect dubious tags.Glider87 (talk) 13:40, 27 July 2015 (UTC)
 * This is correct. The tags put there by Dondervogel are incorrect.Fnagaton 23:37, 27 July 2015 (UTC)


 * The JEDEC say "terabyte commonly used as a prefix to units of semiconductor storage capacity and meaning 240 [1 099 511 627 776] bytes." Fnagaton 23:58, 27 July 2015 (UTC)

Disputed references
I added the {dubious} tag because I am not convinced the references cited are reliable sources that back up the statement. The text containing the disputed references reads: 1 TB is also used in some fields of computer science and information technology to denote 1 099  511  627  776 (10244 or 240) bytes, particularly for sizes of RAM or storage. A JEDEC standards document JESD84-B42 mentions tera in the binary sense "2 Tera bytes (4 294 967 296 x 512B)" on page 86. I propose we discuss the reliability of these references here. Dondervogel 2 (talk) 10:12, 29 July 2015 (UTC)
 * The discussion has implications for other pages, such as Petabyte, Exabyte, Zettabyte, Yottabyte, Brontobyte, Geopbyte and Binary prefix. I will therefore alert those pages (except brontobyte & geopbyte, which do not have WP articles) of the discussion here. Dondervogel 2 (talk) 10:50, 29 July 2015 (UTC)
 * Pages notified. For good measure I also included Timeline of binary prefixes. Dondervogel 2 (talk) 11:01, 29 July 2015 (UTC)


 * Nice attempt at forum shopping. The talk section above contains more than enough reliable sources to demonstrate binary prefix use for tera, peta, exa etcGlider87 (talk) 12:21, 1 August 2015 (UTC)

Use of "TB" for supercomputer memory
Now that the disruptive editors have been moved on, how about revisiting supercomputers. The following list is copy-pasted from an earlier post


 * 1) Hoisie, A., Johnson, G., Kerbyson, D. J., Lang, M., & Pakin, S. (2006, November). A performance comparison through benchmarking and modeling of three leading supercomputers: blue Gene/L, Red Storm, and Purple. In SC 2006 Conference, Proceedings of the ACM/IEEE (pp. 3-3). IEEE. "32 TB" of "Total Memory"
 * 2) Yoo, A., Chow, E., Henderson, K., McLendon, W., Hendrickson, B., & Çatalyürek, Ü. (2005, November). A scalable distributed parallel breadth-first search algorithm on BlueGene/L. In Supercomputing, 2005. Proceedings of the ACM/IEEE SC 2005 Conference (pp. 25-25). IEEE. "32 TB of total memory"
 * 3) Strande, S. M., Cicotti, P., Sinkovits, R. S., Young, W. S., Wagner, R., Tatineni, M., ... & Norman, M. (2012, July). Gordon: design, performance, and experiences deploying and supporting a data intensive supercomputer. In Proceedings of the 1st Conference of the Extreme Science and Engineering Discovery Environment: Bridging from the eXtreme to the campus and beyond (p. 3). ACM. "64 TB DRAM"
 * 4) Preissl, R., Wong, T. M., Datta, P., Flickner, M., Singh, R., Esser, S. K., ... & Modha, D. S. (2012, November). Compass: a scalable simulator for an architecture for cognitive computing. In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis (p. 54). IEEE Computer Society Press. "256 TB memory"

Any objection to citing one or more of these as examples of the use of TB for supercomputer memory? Taken at face value, the most notable ones (from this list) seem to be #2 (earliest publication date, Nov 2005) and #4 (largest memory, "256 TB") Dondervogel 2 (talk) 21:11, 9 August 2015 (UTC)


 * Good idea. I suggest just 2, both in the lede (memory and storage usage) and in Section 1 History, replacing "Motherboard Memory" with "Computer Memory" or other such  category.  Tom94022 (talk) 00:50, 10 August 2015 (UTC)


 * I'm not quite sure but the application is, but similar documentation exists for IBM mainframes. For example, the "IBM z13 Technical Guide"  on page 49 (the section starting the description of the memory subsystem, begins with "The maximum physical memory size is directly related to the number of CPC drawers in the system. Each CPC drawer can contain up to 3,200 GB of physical memory, for a total of 12,800 GB (12.5 TB) of installed (physical) memory per system."  Rwessel (talk) 04:47, 10 August 2015 (UTC)
 * Isn't this a somewhat redundant example of using TB in a binary sense for "Computer Memory" but since the IBM document is dated April 2015 it is not relevant to Section 1. We now have examples in supercomputers, mainframes and servers which should be sufficient to change the article. Tom94022 (talk) 07:14, 10 August 2015 (UTC)
 * Actually the IBM ref is notable for a different reason: it uses "16 EB" in the binary sense. Perhaps something for the Exabyte article. Dondervogel 2 (talk) 21:38, 10 August 2015 (UTC)
 * For now I added the "32 TB" supercomputer memory under 'History'. I'm not sure what to do with the lede.  Dondervogel 2 (talk) 21:50, 10 August 2015 (UTC)
 * The binary use of "16 EB" is in Table 3-6 (p124) in the column "Maximum main storage/Architecture" and rows "z/VM" and "zAware". No idea what the acronyms mean. Dondervogel 2 (talk) 12:45, 11 August 2015 (UTC)
 * In case anyone cares: z/VM is the virtualization system for IBM mainframes (think VMWare, but it's own OS, and doesn't run on top of another OS like VMWare does), zAware ("z Advanced Workload Analysis Reporter") is a diagnostic and management tool. An LAPR is a logical hardware partition, effectively a slice of a machine, in which you can run an OS of some sort.  Sort of a VM-lite.  You can define LAPRs to run in different modes, with "ESA/390/zArch" mode being the "normal" one.  In VM mode, for example, you can assign certain specialty processors, for example IFLs (licensed to run only Linux, and not the tradition mainframe OS's, and costing much less) into a partition, which VM can then manage (IOW, you can then run Linux on a cheap core under zVM).  OTOH, you cannot put an IFL, for example, into a ESA/390 partition.  In other cases they’re specifying the zVM and zAware LPARs (not to mention ESA/390 LPARs) can architecturally be defined with up to 16EB of storage.  The hardware does not support that much, of course, and I don't know if any of the OS's or other things you can run in an LPAR support that either (zOS, for example, is limited to 4TB at the moment, even though you can define a larger partition on a z13 (sort-of the doc says you can’t, but they appear to be taking the limitations of the OS into account), I think zVM has higher limits, but I'm not sure if it goes all the way to 16EB yet).


 * FWIW, Tera/Peta/Exa (and T/P/E) are all used in the binary sense in the "IBM z/Architecture Principles of Operation" (defines the ISA). I was not able to find a online link to the original version from 2000(?), but I did find a -03 (third revision, 2004): .  For example, page 3-34, in the section on dynamic address translation, includes "An address space-control element may be a segment-table designation specifying a 2G-byte address space, a region-table designation specifying a 4T-byte, 8P-byte, or 16E-byte space, or a real-space designation specifying a 16E-byte space. (The letters K, M, G, T, P, and E represent kilo, 2**10, mega, 2**20, giga, 2**30, tera, 2**40, peta, 2**50, and exa, 2**60, respectively.)"  The -01 (2001 - I have a copy) and the -10 (current - also available online) revisions of that manual contain approximately the text.  Pages xx and xxi of -03 also contain explicit definition.  Rwessel (talk) 17:43, 11 August 2015 (UTC)

Merge Kilo/Mega/Giga/Tera/Peta/Exabyte, perhaps to Binary prefix?
I kinda hate to do this, but is there really justification for this raft of small, and overlapping, articles? Perhaps this should be merged into Binary prefix and/or Timeline of binary prefixes? I think even just a single article with individual sections largely covering what the multiple articles cover now would be an improvement. The topics are not really independent. I do see that there was a brief discussion at Talk:Gigabyte a few years ago. Rwessel (talk) 22:02, 10 August 2015 (UTC)
 * I would support a merge in principle, but not to Binary prefix. I quite like the German approach, which is to combine them in Byte. Dondervogel 2 (talk) 22:28, 10 August 2015 (UTC)
 * Agree with one article, perhaps "Byte", not Binary prefix, with appropriate redirects therto for the specific prefix combinations. Tom94022 (talk) 06:04, 11 August 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Terabyte. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive http://web.archive.org/web/20110718233004/http://www.dkrz.de/pdf/poster/ISC10-Poster_Web/ISC10_HardwareDKRZ.pdf to http://www.dkrz.de/pdf/poster/ISC10-Poster_Web/ISC10_HardwareDKRZ.pdf

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 14:55, 27 February 2016 (UTC)

Misleading redirect from "tbyte"?
Wiktionary says that a "tbyte" is an 80-bit (10-byte) integer, not a terabyte. Equinox ◑ 14:39, 8 January 2017 (UTC)
 * sounds like an error in Wiktionary to me Dondervogel 2 (talk) 15:41, 8 January 2017 (UTC)


 * Searching around suggests that TBYTE was a ten-byte data structure in IBM assembly language or something. Probably more often does mean terabyte now. Equinox ◑ 19:54, 8 January 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Terabyte. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20090508143215/http://www.ygroupsblog.com/blog/2009/03/17/groups-search-update to http://www.ygroupsblog.com/blog/2009/03/17/groups-search-update/

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 19:34, 26 July 2017 (UTC)