Talk:Cloud computing

Vertical Market Removal?

The Vertical Market section was removed. I would like other editors to comment on this rather than leave one editor to make this decision. My intention was this was going to expand into different verticals with links to different articles. I request the opinions of other editors on this issue before I restore it.--Ourhistory153 (talk) 17:52, 18 December 2011 (UTC)

Technical Gibberish

This article is written by a technician for other technicians--not for an ordinary consumer-user. I counted more than 50 terms that I was unfamiliar with, and several areas that I thought I didn't need to know to understand what the "cloud" is (my original intent for accessing this page in Wikipedia.

The first paragraph should start out with an easily digestible definition of what the cloud is and how it affects individual users. Then, that first paragraph should contain a bulleted list of what follows.

The rest of the article should be edited for relevance to end users and grouped accordingly: ordinary readers such as myself/extraordinarily-aware groups of users. Language has to be simplified in the areas not targeted for the more knowledgeable group; and if the "ordinary readers" group has almost nothing here for it--as I think might be the case--then relevant areas of the text should be adapted for it. (British English should be left in--we understand it, after all, and this article has far worse things going on with it.)

I didn't understand what the "cloud" was before I accessed this page, and felt so talked down to, that I never did realize my goal at the end.

(Sudarat64 (talk) 22:11, 9 January 2012 (UTC))

The criticism is valid. I don't believe the article deserve a B rating. I have taken it down a notch to C. I will try to find time to give this some love. --Kvng (talk) 17:01, 11 January 2012 (UTC)

There is now an generally accessible Introduction to cloud computing article, which I believe overcomes these criticisms.

Fcalculators (talk) 02:51, 14 March 2012 (UTC)

I gave this article only two stars for objectivity, since there isn't a lot of it to be found—the entire article reads like sales literature. In my mind, cloud computing is another synonym for snake oil. The concept stinks from a security standpoint. "Lessee, we'll give control of our data to some unknown entity, who may well steal it for nefarious needs." Sounds pretty stupid to me. If you don't own it you can't control it.

Bigdumbdinosaur (talk) 01:55, 16 March 2012 (UTC)

Gibberish - yes, Technical - hardly. As pointed above, whole article reads as sales add, probably because "Cloud computing" is not a technical term, but fashionable weasel word invented by marketing people trying to rebrand decades old concepts as something revolutionary.

Unfortunately it is in frustratingly common use usually by non-tech personnel in tech meetings, presentations etc. Unsurprisingly when asked to elaborate and be more specific, we usually get either more weasel words or description of simple web service. — Preceding unsigned comment added by 210.61.64.216 (talk) 07:22, 27 March 2012 (UTC)

Minor Content Mis-match - Servers and the Cloud

Just a minor point...

The current diagram shows "servers" outside the cloud. However, this referenced content statement within the article - Cloud computing extends this boundary to cover servers as well as the network infrastructure.[28] - indicates "servers" are now considered within the cloud. Perhaps a revised diagram could resolve this "inconsistency". I suggest something showing servers both internal and external to the cloud since servers, at times, are like as other "user access devices" and are also cloud components...along with a brief explanation.

Nealct56 (talk) 15:33, 17 January 2012 (UTC)

Towards an improved intro for this article

Generally, the intro should be shortened and made more concise. Here are some specific points:

• "Cloud computing is a marketing term": Well by now, it is more than a marketing term. It is reality in many peoples lives and it is big business.

Done --Bikeborg (talk) 13:55, 30 January 2012 (UTC)

• (I like the analogy with the electricity grid. Makes the concept better to grasp for non-experts.)

Somebody wiped it out. I will eventually restore it. --Bikeborg (talk) 16:40, 9 February 2012 (UTC)

• Reference 4 "cloud computing defined" appears to be a scam to generate clicks on somebody's web page. The cloud computing definition should be here on Wikipedia and not elsewhere. Should be removed.
• The remarks on thin client applications and screen sharing are too specific for the introductory part of the article.

Maybe we need a separate section where we can go into more detail about certain cloud aspects. Maybe into the Characteristics section?

• I am not familiar with the term "converged infrastructure". Is that a Wikipedia notable concept or a marketing term?
• Reference to US government initiative: There are many many cloud initiatives in business and administration across the world. We should not discuss them here in the introduction.

I was bold and removed it. Please let me know if you disagree --Bikeborg (talk) 13:55, 30 January 2012 (UTC)

--Bikebot (talk) 11:05, 25 January 2012 (UTC)

Good list. I have replaced the unreliable ref. Converged infrastructure is definitely notable. --Kvng (talk) 14:59, 27 January 2012 (UTC)

The more I study the article the more it appears to me to require an overhaul. For example:

• The characteristics section talks more about the benefits than the characteristics. Should we separate benefits (value prop) and characteristics into two sections?
• Section 4.1: Looks like a Google afficionado entered the examples. Seriously: Devices specifically designed to be cloud clients are still the exception. Most devices are multi-purpose. I will try to clarify the policy on product examples.(see Wikipedia:spam)
• Section 4.5: Server: Isn't the server part of the infrastructure layer? As well as the network?
• The list of references needs a major review. There is a lot of them that are more incidental than fundamental reference literature.

--Bikeborg (talk) 13:55, 30 January 2012 (UTC)

Revision of Layers Section

I am currently working on a revision of the Layers section. The main problem I see currently is that the current text does not prominently reflect the Service Models (IaaS, PaaS, SaaS) that everybody talks about and that are also defined in the NIST paper.

Also, cloud client and cloud servers are different beasts from the service models.

Therefore, I propose to do the following changes:

• Make one section called Service Models describing IaaS, PaaS, SaaS
• Make a separate section about cloud clients.
• Create a new picture

I am working on a draft text here - but it will take some time. Any help appreciated.

--Bikeborg (talk) 16:03, 9 February 2012 (UTC)

Introduction to Cloud computing

Introduction to Cloud computing popped up on a new article list, and whilst I appreciate the huge amount of work that has been put into it, I think it might infringe the WP:NOTGUIDE or WP:NOTTEXTBOOK guidelines. The title alone seems wrong for an encyclopedia. But as this isn't my area of expertise, I thought I'd raise it here for those more involved to decide whether it's a useful article or not. -- The-Pope (talk) 16:18, 13 March 2012 (UTC)

I posted the Introduction to cloud computing article in order to improve the Wikipedia content for this subject. As you can see if you read the past discussions, there have been many complaints about the main article, and often from an understandability or end-user point of view (such as the Technical gibberish comment above). I have called for improvements to (or even a re-writing of) the article, and I was responsible for the call to have it downgraded to a C rating. In the discussions, I suggested that one way forward was to write an introductory article. There are many such articles (you can see all/many of them by typing "Introduction to" or "Introduction to a" - or b, c etc - into the search box). The article has been edited by a member of the computing group, and described by that editor as a "generally accessible, non-technical article", so it seems that others accept it.

I still believe that the current article needs much improvement, and because of the introductory article, I believe it can be more openly technical. Fcalculators (talk) 03:09, 14 March 2012 (UTC)

I don't assume we need an introductory article. The topic is not all that technical. The problem here is that the current article needs work. I think it can be improved by incorporating this new work into it and deleting a bunch of the difficult to understand material. I have put up merge banners. --Kvng (talk) 21:01, 15 March 2012 (UTC)

I agree that the current article needs work, but I believe that the subject can be quite technical, especially if the article explains how it all works by dealing with the architecture of cloud datacentres, and the implementation technologies. Also, I believe it would be beneficial to cover the following from a more technical point of view: all types of service, deployment and typical characteristics, with specific diagrams; the various ways that providers can operate, using in-house resources and/or other service providers (as evidenced by comments in the discussion archives, the layers diagram can confuse some readers by implying that each layer must be built on top of the ones below); details of specific examples, rather than just one or two-sentence descriptions; P2P versus client-server examples; the NIST Reference Architecture.

This is a top-importance, extensive subject with a large and wide audience of technical and non-technical readers, and with aspects ranging from simple facts to in-depth technologies for consumer access and provider implementations. Also, the examples of cloud computing cover the whole range of activities, professional and personal. So I can’t see that a single article can cover all of this adequately, without being too long or difficult to follow for many readers.

My idea with the introductory article was to deal with the basics, especially for the non-technical section of the audience, so that the cloud-computing article itself could then be more technical without end-users complaining that they couldn’t understand it. Fcalculators (talk) 01:30, 20 March 2012 (UTC)

Introduction? Really? It is longer than this "main" article, and, it did not seem to be any easier to figure out than this. 85.217.20.177 (talk) 09:33, 20 March 2012 (UTC)

There's no problem that it's longer than the main article, since it's an "introduction to the topic", not "the introduction of the article". As Fcalculators points out, they have different audiences with different needs; the Introduction article should include a general description of the field, and the main article a detailed explanation of the primary technical concerns. "Introduction to..." articles are common for several high profile topics (we even have a template for it). As long as it doesn't try to teach "how to build your own cloud service" and keep itself with defining the basic terms, it should be fine. Diego (talk) 13:04, 20 March 2012 (UTC)

• Support merge I would support a merge, just so that a central high quality item can emerge. In terms of overall quality these two are actually better than many of the other ProjComputing articles out there. But it would still make sense to have a really nice item as a show case of how Wikipedia can actually achieve scholarly status. However, I did oppose merge of Cloud computing security into here. History2007 (talk) 23:02, 20 March 2012 (UTC)

• Support merge. I don't seem to grasp the logic of an Encyclopedia holding a universe of articles plus another parallel universe of "Introduction To..." articles on the same subjects. It would be not only unnecessary, it would be simply ridiculous. --AVM (talk) 16:37, 21 March 2012 (UTC)
• Support merge BUT This article is already uncomfortably long, and merging the very valuable content from Introduction will push it beyond practical. Likely we will need to split off bits and pieces soon anyway, and some elements of Introduction might then rate their own, something like specialised language of cloud computing, &c. YamaPlos talk 19:11, 27 March 2012 (UTC)

Private Cloud statement "must buy, build, operate" is not true...

Private cloud infrastructure CAN BE and IS procured "as a service" that is billed monthly as a utility and based on allocated capacity. Utility billed private cloud infrastructure is currently in operation in several Federal government agencies, and originated in the Defense Information Systems Agency (DISA). DISA is the Department of Defense Combat Service Agency that provides centralized enterprise IT and global communications to the U.S. Military and other Defense Agencies). Payments are made monthly to capacity services vendors and include all maintenance, support, and tech refresh.

Additional users of "capacity services" IT infrastructure as a service offerings include, at the least, the U.S. Department of Agriculture (USDA), the Air Force Medical Support Agency (AFMSA), the Department of Homeland Security (DHS), and the National Institutes of Health (NIH).

HISTORY DISA began procuring mainframe and client server computing capacity as a service in 2001 and 2003, respectively. Storage as a service began in 2003. The 2001 acquisition was to Unisys for Unisys mainframe and storage. In 2003, the acquisition vehicle was called ACE, for Assured Computing Environment, and included IBM mainframe equipment, plus storage.

In 2005, several multi-year "Server Processor Environment" contracts were awarded by DISA for on-demand compute capacity as a service. Winning vendors included: SUN, Hewlett Packard, ViON Corporation, and Apptis. The award of these contracts was discussed in this article: http://gcn.com/articles/2006/10/24/disa-on-demand.aspx

In 2007, storage as a service was procured on its own multi-year contract, Enterprise Storage Services (ESS), which is discussed in this article: http://www.military-information-technology.com/mit-home/367-mit-2011-volume-15-issue-10-november/4995-storage-on-demand.html

Note: the vendor-owned equipment provided under the Enterprise Storage Services contraact provides the "private cloud storage" that is included in the DISA RACE DoD Private Cloud offering. The DISA RACE private cloud was launched in 2008. "RACE" stands for Rapid Access Computing Environment. RACE features user-self provisioning for highly virtualized client server computing (X-86, Linux, Solaris) and enterprise storage capacity on demand in a cloud offering.

The initial Federal breeding and testing ground for the "capacity services" (capacity on demand) method to acquire IT infrastructure as a service was the DISA data centers (called "Defense Enterprise Computing Centers" or "DECCs"; and "Tactical Enterprise Computing Centers" or "TECCs"). The DISA DECCs are built and operated at Tier IV+ standards, and DECC-equivalency is the defacto standard for U.S. military and defense agency data centers. Discussion about what constitutes "DECC-equivalency" is not widely discussed in unlassified forums, however reference to "DECC equivalent" sites can be found in this unclassified DoD budget document: http://www.dtic.mil/descriptivesum/Y2006/DISA/0303170K.pdf

These recurrence of Federal government acquisitions of computing and server infrastructure, plus the addition of a new network equipment "as a capacity service" has proven across many years that the use of vendor-owned equipment provisioning computing, storage, and network infrastructure as-a-service inside government-owned and operated data centers is cost effective and significantly lowers risk of security breach. In fact, security breaches were so effectively negated that in 2011 the Department of Defense determined the DISA DECCs to be more secure and cost effective hosting environments than the Department of Veterans Affairs (VA) can provide for itself, and the VA began using the DECCs to host its data. This is significant because the CIO of the VA must report to Congress on a quarterly basis its status in protecting against additional data breaches, following the 2006 data breach whereby 26.5 million records were inappropriately downloaded from the VA.

IN CONSIDERATION of the facts and examples detailed above, involving highly credible customers (U.S. Federal government agencies; large and well-regarded multi-national technology manufacturers as well as innovative small businesses) and occuring for many years, I request that you correct the article to indicate that Private Cloud infrastructure is, in fact, available in on-premise, on-demand, and billed as a utility.

Thank you in advance! — Preceding unsigned comment added by LaFemmeTech (talk • contribs) 18:27, 13 March 2012 (UTC)

Cloud computing ancestor.

In the early sixties, MIT's Project MAC aimed at provide shared computing resources at a time when one who needed computing power had to record his own programs and data on punched cards. In these days the most common, not to say unique, communication tool was the TTY teleprinter.

Unhappily the first experiments of on-line computing, by linking a TTY to a computer, accepted a very limited number of simultaneous users. The reasons were the low performance and capacity yielded by the hardware available in the sixties.

Nonetheless, from an architectural design standpoint, Project MAC outlined a general computing environment, including a high degree of security/privacy and a real ability to take advantage of more powerful hardware to come.

MULTICS systems, designed and engineered by General Electric, have represented a milestone in the project development. Bell Laboratories operated several of these machines for their computer developments, including UNIX (allegedly named after "MULTIX"), as General Electric/Honeywell did for their software factories (PL1 being the main working language).

On the other hand, General Electric made commercial offers for "utility computing services" based on the same platforms, by 1968.

Astonishingly, the concept and the services they marketed were similar to those provided by the present "cloud computing".

"Independent" personal micro computers were by large more in line with the mood of that period, and Utility computing did not emerge as a success ! A simpler version, Time-sharing, had been a little more successful.

Bernard Huet193.248.13.239 (talk) 13:28, 18 March 2012 (UTC) 193.248.17.158 (talk) 21:28, 15 March 2012 (UTC) — Preceding unsigned comment added by 193.248.13.239 (talk)

History: AWS Launch Date

Article claims: " launched Amazon Web Service (AWS) on a utility computing basis in 2006" which leads to a conclusion that AWS was launched in 2006. However, AWS was launched in July 2002, as noted on the Wikipedia AWS page (http://en.wikipedia.org/wiki/Amazon_Web_Services) the source of which is Amazon's mediakit timeline, undoubtedly reliable source: http://phx.corporate-ir.net/phoenix.zhtml?c=176060&p=irol-corporateTimeline

The sentence should be reworded if it meant some more specific connotation and some sort of clarification about 2002 date must be added. In current form article leaves an impression that first commercial cloud service (AWS) was offered 4 years later than it actually was. That's no small difference.