Talk:Hardware virtualization

Reverted Definition

Trevor, I do not doubt your integrity or intentions, but you replaced the previous definition (mine) with an unverifiable opinion that is extraordinarily broad ("virtualization = abstraction"), with no reference or source to back it up. I have published many articles and interviews on this topic in both computing and mainstream press, and have many other credentials that I do not care to share with the world at large. I added the definition that you deleted because I was professionally appalled at how poor the previous one was. I believe that the new description is much worse. I referenced the definition from an independent IT analyst firm, so how can you justify the accusation of "vendor spam?" In any case, it was clear, valid, and meaningful. So I have reverted the definition, within the context of your edit.

Frankly, I think a lot of this article needs further revision. It contains vast amounts of at best poorly cited and unverifiable opinion, and much of it is simply quite wrong. The over-simplified two-tiered taxonomy appears to be at best a personal construct - unverified, unreferenced, and unreflected in any mainstream thought. Other definitions, history, and examples are both incomplete and inaccurate. There is also too much vendor bias here as well. I do not currently have the time, but would like to clean it up significantly. It remains far from Encyclopaedic.

(Not a registered user - and not a vendor either, thanks very much - just an honest and concerned IT pro). 67.190.156.208 09:29, 31 December 2006 (UTC)

A few points:

• I have no argument about the need for revision here. There are a number of related articles on this and related topics, all needing attention. Some of the others are in worse shape.
• I have not had time to assemble good source material for these articles, nor to tackle their restructuring/rewriting – which I agree is needed. There are a number of good early sources cited in other articles I have worked on recently, e.g. History of CP/CMS and IBM CP-40; these articles are referenced on this page, but I did not cite their sources here.
• As far as "vendor spam" is concerned: I expect you would agree that, if an EMA employee had supplied this quote, along with the link to EMA's website, this would appear to be a form of advertising. This is a common problem at Wikipedia, and that was indeed what I guessed had happened – particularly given the anonymous source. I do tend to be suspicious of anonymous contributors – no offense meant.
• Quoting a definition from a consultant's website does not strike me as encyclopaedic, even if the quote is good. It would make sense to use a good quote from a well-known author, writing in a mainstream academic publication. Otherwise, I think the Wikipedia approach would be to write a new description, and cite the most credible sources available to back it up. Given your professional credentials, I'd think this would be your preferred approach. I am NOT challenging your re-insertion of the quote, though I might have problems with a vendor link (and how can we use the quote without a link?). At any rate, I hope you can see how it struck me when I first saw it – like all the Virtual Iron links that keep showing up everywhere.

Thanks for explaining your views. I do regret that you choose not to be a registered user, which adds credibility through continuity. You can still hide your identity, if that's the goal. Trevor Hanson 21:38, 31 December 2006 (UTC)

User:Kubanczyk added "unreliable source?" to the definition, but didn't explain why. In the history s/he just said "I don't like this definition". But just because you don't like a definition doesn't make it bad or unreliable, and just saying the source is unreliable without replacing it with a reliable source seems pretty pointless. And actually, that definition is in my university e-business textbook, so I added it as a citation, because that seems pretty reliable. I did some looking round and found that this is rally a pretty common definition, so I added more citations from a lot of other places that seemed reliable too. I thought that it was probably helpful but I am not exactly a pro Wiki editor, so if this is not, please comment here and tell me what would be better.

Thank you, now it's better. Old source was unreliable for me because it was neither easily available via url provided, nor it was a well-established book. Such warning sign from an accidental non-export reader (i.e. me), is far from being "pretty pointless". In fact it is a standard Wikipedia's practice, and a core of wiki philosophy.

Now, the definition. The problem with old definition was that it hardly defined anything. It just said that "virtual" is another word for "logical", but left "logical" undefined. This was not acceptable for me. I like a new definition after your edit, it is much more informative now. --Kubanczyk (talk) 10:17, 22 April 2008 (UTC)

Changing en dashes to em dashes

User:Bovineone just did an edit changing lots of en dashes to em dashes, citing WP:MOS. However, I might point out that MOS says (emphasis added):

As with any guideline that permits multiple styles, editors are encouraged not to convert others' dash styles without good reason....

The following five dash styles are currently in use on Wikipedia:

• Tight (unspaced) em dashes—like this. Entered by means of —, or, if your keyboard allows it, you can type it in directly, or you can retrieve it from below the edit window (as discussed above).
• Spaced em dashes — like this.
• Spaced en dashes – like this. Entered by means of –, directly if your keyboard allows it, or retrieved from below the edit window. (Note: an unspaced en dash is properly used to indicate a range of numbers; unspaced en dashes should not be used for the parenthetical or colon-type uses, as discussed above.)

etc.

In other words, the spaced en dash usage that had been there was a) considered normal and b) a precedent for this article. I haven't reverted this edit, but I don't think that change was following the Wikipedia way. Trevor Hanson 21:09, 16 January 2007 (UTC)

en-dash is more commonly used for ranges and compound terms, while em-dash is more appropriate for separation of thought. The intent of the section that you quote is primarily intended for converting between uses that have equivalent established meanings, and not as much for incorrect usage. In any case, the definition syntax is probably a better choice anyways, so I've change it to that. -- Bovineone 21:40, 16 January 2007 (UTC)

Remove some of the links which are obvious product promotion

and added in a link to QEMU

81.174.174.31 18:03, 6 April 2007 (UTC)

'Blatant self-promotion'

I reverted a 4/18 change (to 'remove blatant self-promotion'). That change had removed the source of the quote in the overview. I don't like the quote either, and had removed it earlier, but this led to a small exchange (see above at Talk:Virtualization#Big_edit_on_12.2F8.2F2006 and Talk:Virtualization#Reverted_Definition). I have put the source back in on the quote (but in a reference this time), because having a quote without a source is pointless. But by all means replace the quoted description with your own, better one. (Shouldn't be too hard.) I backed out of this issue at the time, since it seemed to be verging on an edit war. Trevor Hanson 00:46, 19 April 2007 (UTC)

Opposite of Transparent

I think the concept is not quite right and very wide. Virtualization in IT area is referred to something which do not really exists, but is clearly visible, perceivable. The opossite counterpart is transparent, which refers to something invisible but real, something that exists in fact. Appliable to Glasnost, transparency_(computing), virtual machine, virtual reality, virtual friend, virtual sex. I add that concept, but i would rather like this to be the main concept. Up to the owner. --Rodolfoap 00:01, 25 May 2007 (UTC)

"Up to the owner": FYI, there is no "owner" in Wikipedia; or rather, all editors are owners. Trevor Hanson 03:36, 25 May 2007 (UTC)

Not sure if this is quite the same thread, but I'll put it here. I came to this page wondering what all the excitement about virtualization was; I came away mystified. That is, I kind of understand what it *is*, but I didn't understand *why* you would want to do this.

Until I went to the VMware page, where I read this:

The host provides pass-through drivers for guest USB, serial, and parallel devices. In this way, VMware virtual machines become highly portable between computers, because every host looks nearly identical to the guest. In practice, a systems administrator can pause operations on a virtual machine guest, move or copy that guest to another physical computer, and there resume execution exactly at the point of suspension. Alternately, for enterprise servers, a feature called VMotion allows the migration of operational guest virtual machines between similar but separate hardware hosts sharing the same storage area network (SAN).

That explains at least part of the motivation; maybe there are other reasons? (such as higher utilization of an underutilized piece of hardware?) And does the above apply to all types of virtualization?

It's also unclear to me after reading this what the difference between virtualization and the old notion of a terminal server is, particularly in the bad ol' days when a terminal was the only way to log into a computer, and thereby "see" its OS. Is virtualization terminals on steroids? 72.66.121.13 (talk) 14:41, 29 June 2008 (UTC)

Your confusion is partially due to the fact that when people say "virtualization" they more often than not mean "virtual machine technology". And I'm not surprised that the article didn't help you much, it doesn't have a expert-subject tag for nothing. It's all over the place and needs a major re-write. Heiser (talk) 17:31, 29 June 2008 (UTC)

Platform virtualization involves the simulation of virtual machines

I've modified the above sentence to "Platform virtualization involves the simulation of computer environments"; Virtual Machines are the simulation not what is being simulated. The analogy i like to draw is to a flight simulator; the virtual machine is the flight simulator and the computer environment the aircraft.

Request to link to this page

I'm a new Wikipedia contributor. I've written a page on network virtualization that is basically an overview and summary of network virtualization examples. The page is listed under the virtualization software category page. I'd like to link to your Virtualization page as well. I'm not 100 percent sure how to do this, as I've just begun to make Wikipedia contributions.

Stefaniab 20:04, 8 October 2007 (UTC)

Hardware enabled virtualization

Hardware enabled virtualization - some comments:

 the virtual machine has its own hardware and allows a guest OS to be run in isolation.

Has nothing to do with Hardware enabled yet. not very informative.

 In 2005 and 2006, Intel and AMD provided additional hardware to support virtualization. 
 Examples include VMware Workstation, VMware Fusion, Microsoft Virtual PC, Xen, Parallels Desktop for Mac, and Parallels Workstation.

how are these examples for the additional hardware to support virtualization.
Is it fair to suggest that we remove this paragraph as long as it does not provide any information nor being logical ?

Weiss gal wiki (talk) 21:42, 27 January 2008 (UTC)

Portable Applications

I think that the line

 It is unclear whether such implementations are currently available.

should be removed from this section as it is abundantly clear—at least to me—that Thinstalled applications are an example of exactly this type of implementation. I'll check back for discussion in a day or two and if no one has issues I’ll take it out. I am not yet quite at the level that I would consider myself an ‘expert’, but I am teaching a university course on operating system installation this term and as I go through teaching the course I hope to contribute to making the sections on virtualization, desktop virtualization, virtual machine and perhaps a few other a little more rigorous and (hopefully) more usable in an academic as well as a business context. (The best way to learn something really well is to teach it...) Ray Trygstad (talk) 21:08, 28 January 2008 (UTC)

Needs more on VMWare

I came here expecting to find a page I could refer people to about virtualisation in general, and pointing at the key role in that now played by VMWare.

I expect we're being overly shy of referring to products. Sometimes what looks like advertising, just isn't. Andrew (talk) 06:56, 1 February 2008 (UTC)

Splitting article

I've just rewritten the intro to be reasonably readable by normal human beings, but it doesn't help that, as was pointed out earlier on this talk page, the article currently uses a ludicrously broad definition of "virtualisation" which was apparently extracted from a first-principles examination of the definition of "virtual". I believe that this article should focus exclusively on what the article currently terms "platform virtualisation", punting "resource virtualisation" to resource abstraction (as "abstraction" is a far more common name for this principle at present). Chris Cunningham (not at work) - talk 09:41, 26 April 2008 (UTC)

I agree that a lot of this article should be moved to Platform virtualization (in WP:Summary style), but the article Virtualization should stay and cover, on a very academic (!) level, in a short way, without mentioning any commercial examples, both platform and resource virtualization. The main resource virtualization topics that a reader will expect here are: memory (see Virtual memory), disk storage (see Storage virtualization), tape libraries (see Virtual tape library), network interfaces (vmWare or VIOS functionality). I don't think "resource abstraction" is ever applied to these, and by all means it is not a WP:COMMONNAME. Those topics are different from each other, so it is OK to leave the current section - a brief list of wikilinks. --Kubanczyk (talk) 10:24, 26 April 2008 (UTC)

If that's the way to go (and I believe you're right), I think the setup should be:

1. A page at virtualization for platform virtualisation (because the most common term in the vernacular for this concept is definitely just "virtualization")
2. an {{otheruses}} template at the top of said article pointing to:
3. virtualization (disambiguation), a series of links to the various things termed "virtualisation" which aren't platform virtualisation

This seems to address all the current issues. Chris Cunningham (not at work) - talk 10:30, 26 April 2008 (UTC)

What I've tried to express is an weak oppose to your 1. "A page at virtualization for platform virtualisation". For me this is not WP:COMMONNAME. Such article would need some very good sources, to back up the statement "virtualization today turned out to generally mean platform virtualization".

A quick examination of the article's current references shows that they use the term "virtualisation" almost exclusively to refer to what the article currently deems "platform virtualisation". A quick Google for "virtualisation" indicates that almost all references to "virtualisation" are to platform virtualisation. This is basically the embodiment of WP:COMMONNAME. Leaving virtualisation as a disambig page lends undue weight to the minority terms. Chris Cunningham (not at work) - talk 13:52, 28 April 2008 (UTC)

But virtualization clearly includes network virtualization, desktop virtualization, presentation virtualization, storage virtualization, cluster computing, file virtualization, memory virtualization, and several other forms of computing abstraction, and we can't say they aren't virtualization, just because one type ("platform virtualization") is more popular (i.e. most hits on Google, ~= more marketing $$ by software companies) than the others.

I would suggest something similar but a little different:

1. A page at virtualization with just the existing broad definition and introduction
2. an {{otheruses}} template at the top of said article pointing to:
3. virtualization (disambiguation), a series of links to the various things termed "virtualisation" (i.e. including platform virtualisation)

This would give a casual reader the chance to understand the broad concept without the additional confusing clutter, while deeper readers can click through to any of a number of specific topics (which can also be cleaned up individually!). 65.74.240.100 (talk) 04:09, 30 April 2008 (UTC)

Actually you don't need a separate virtualization (disambiguation). I think a decent alternative to Chris'es proposition is:

1. A disambig page at Virtualization with broad definition and list of links (i.e. including Platform virtualization)
2. Most of other contents of the article moved to Platform virtualization (or other WP:COMMONNAME)

I agree that "v=pv" is just the current trend, and judging on Google/Wikipedia is too much WP:OR in this case. As soon as WP:RS are provided, I will withdraw.

--Kubanczyk (talk) 08:29, 30 April 2008 (UTC)

The problem with this is that it fails to take into account that the majority of wikilinks from other articles to virtualization will be inteded to point at the platform virt article. So for the sake of making links obvious, we should just put the platform virt article there. There's no malice in putting the general concept on a page with "disambiguation" in the title. Keeping the disambig at virtualization seems to be an attempt to use it as a dictionary definition, which really isn't necessary. Chris Cunningham (not at work) - talk 08:58, 1 May 2008 (UTC)

But that is spurious, for the same reason the many Google links justification was. Just because lots of wikilinks assume "v=pv" doesn't make it true, nor is it necessary (or even a good idea) to reinforce the false notion. I do like Kubanczyk's solution immediately above, and still can't see anything wrong with it. But making the Virtualization article the same as Platform virtualization seems to just entrench this misconception for no justified reason. 24.8.144.134 (talk) 04:14, 4 May 2008 (UTC)

It isn't an endorsement of the term, it's simply a reflection of its popularity. Please read WP:NAME thoroughly. As it stands, I think this is a sensible way to move forward, and will hopefully get on with the splitting soon. if it so happens that there's stronger opposition to the exact naming conventions chosen, it can be taken to WP:RM once complete - then at least we've had some progress. Chris Cunningham (not at work) - talk 15:35, 6 May 2008 (UTC)

Wikipedia using a term inherently endorses such usage. Popularity is not fact. Baseless autocratic activity is not progress. There is no WP:RS to support your intention to redefine virtualization as platform virtualization. And you might want to read up on many different Wiki conventions yourself before you unilaterally redefine a page title, such as:

- WP:NAME - "do not write or put an article on a page with an ambiguously named title as though that title had no other meanings".

- WP:NAME - "a term that may be used to describe several different search terms may require a disambiguation page.".

- WP:NCON - "Wikipedians should not seek to determine who is "right" or "wrong", nor to attempt to impose a particular name for POV reasons".

- WP:RM - "If there has been any debate about the best title ... or if anyone could honestly disagree with the move, then treat it as controversial" and "take all three steps listed ... when requesting a move"

- WP:PRECISION - "If a word or phrase is ambiguous, and an article concerns only one of the meanings of that word or phrase, it should usually be titled with something more precise than just that word or phrase".

- WP:D - "Always reach consensus before attempting the split. Wikipedia:Be bold in updating pages doesn't apply, as it is very difficult to revert a split, often requiring extensive assistance by administrators.".

By all of these conventions you are not just wrong in fact, but also wrong in process. But I do not intend to wikilawyer here. My more general point is that pointing to an entire convention is not a valid argument. It is no more than another spurious argument, this time an appeal to authority. It is also arrogant, patronizing, and marginally insulting. Can you instead point to the exact item in WP:NAME that supports your case, rather than just blithely citing its authority as though that somehow proves your point? Then perhaps we can discuss it sensibly.

I have no problem with the idea that this article should be split. But still you have not explained what you think is wrong with Kubanczyk's suggestion, which at least two of us agree on. Rather, you continue to assert only that you know best. As it stands, despite multiple opposition, and with no actual argument except that you (and you alone) think it is 'sensible', you are intending, by changing the fundamental content, to effectively redefine a canonical page based only on WP:OR. Your attitude that you will do it anyway regardless of a lack of WP:CON, and anyone that disagrees can go to (the backlogged WP:RM) hell, is unfortunate. I hope we can instead come to some consensus that achieves our objectives. 24.8.144.134 (talk) 07:22, 7 May 2008 (UTC)

The nutshell summary at the top of WP:NAME is clear enough: "Generally, article naming should prefer what the greatest number of English speakers would most easily recognize, with a reasonable minimum of ambiguity, while at the same time making linking to those articles easy and second nature." Wikipedia:Naming conventions#Primary topic also seems to favour my suggestion because it is a "well known primary meaning". But as this has generated the usual apoplexy about unilateral action against the will of the people, etc., I'm happy to go with the other plan just now and take it to WP:RM afterwards if that will satisfy people. Chris Cunningham (not at work) - talk 07:50, 7 May 2008 (UTC)

Thanks Chris, that certainly sounds like a solid plan to me. 24.8.144.134 (talk) 04:11, 12 May 2008 (UTC)

---

I've now carried out Kubanczyk's suggestion. Chris Cunningham (not at work) - talk 11:00, 6 July 2008 (UTC)

Intel Core computer chips

What is the most significant Intel Core chip being installed in Lap Tops and in Desk Tops? —Preceding unsigned comment added by 24.148.239.215 (talk) 17:41, 13 June 2008 (UTC)

simulating a different instruction set

This article currently mentions "In full virtualization, the virtual machine simulates enough hardware to allow an unmodified "guest" OS (one designed for the same instruction set)".

However, this article fails to mention the situation where the virtual machine simulates a system with a different instruction set -- is there a generic name for that? Is there a generic name for things like Virtual PC for Mac and PearPC ? If such things are technically not "platform virtualization", I think this article should point out that fact and link to some other article about that kind of thing. --68.0.124.33 (talk) 03:09, 13 May 2009 (UTC)

Renaming (yet again)

SamJohnston's renaming from "platform virtualization" to "hardware virtualization" is total nonsense IMHO. Why do people so such things without discussion?

Hardware virtualization would include any virtualization of hardware. This includes virtual memory, virtual disks, lots of other stuff. "Platform virtualization" tried to capture the fact that it refers to all the underlying hardware. However, this isn't the right term either. The term well-established term in the scientific literature is "system virtualization", see Smith and Nair, the architecture of virtual machines. An encyclopedia isn't supposed to invent new terminology, but explain the established. heiser (talk) 06:23, 10 November 2009 (UTC)

Would someone please do me the favour of reverting that stupid renaming, or, even better, rename to the proper term "system virtualization". I don't have the time to fiddle with this. heiser (talk) 20:45, 12 November 2009 (UTC)

Well, regardless what title we choose for this (plain "virtualization" was used at some point), the article on hypervisor also has a lot of redundancy with this one. Except for the classification there isn't much else that's not covered in both. As a first step, I'm merging partial virtualization and full virtualization in here because those are merely concepts with little potential for further expansion. For comparison, we don't have sub-articles for system virtual machine and process virtual machine (I created redirects to sections though, and that's a good thing). Hardware-assisted virtualization is also in poor shape because it doesn't really contain what it should: the mechanisms that help virtualization like interpretive-execution facility, tagged TLBs, second level address translation, etc. I'm currently working on x86 virtualization, as I know more about that, but many hardware mechanisms were simply borrowed from IBM's decades old stuff. By the way, I congratulate you for the work on the virtual machine article from late 2007; it made it much better (following Smith's book definitions). I'll see if I can come up with some recent and hopefully standardized defs and terminology here. Pcap ping 19:14, 21 April 2010 (UTC)

The "Virtualization for Dummies" (free pdf from AMD/Sun) "book" calls this "hardware emulation virtualization", because they wan to distinguish it from they call "operating system virtualization" (aka containers like Solaris Zones, User Mode Linux, Virtuozzo/OpenVZ), pretty much coinciding with Wikipedia's Operating system-level virtualization; I'm guessing these qualify as partial virtualization with the definitions in this article. So it's probably way confusing to rename this article to "system virtualization", but "hardware emulation virtualization" is a mouthful. Pcap ping 19:37, 21 April 2010 (UTC)

This book uses "computer system virtualization" and "full system virtualization" interchangeably. Pcap ping 19:51, 21 April 2010 (UTC)

This Hyper-V book uses "operating system virtualization" (for hw virt), although with Hyper-V the concepts aren't incredibly separated beacause of their extensive paravirtualization when running MS OS on Hyper-V (VMBus etc.) VMWare at least calls its VMware VMI paravirtualization... Pcap ping 20:00, 21 April 2010 (UTC)

Another Hyper-V book also uses "operating system virtualization" (so it's probably MS mandated terminology) interchangeably with "full-system virtualization". Pcap ping 20:03, 21 April 2010 (UTC)

This ESX book calls it "hardware-level virtualization", but really defines only type 1 hypervisors (well, ESX is a type 1). Pcap ping 20:07, 21 April 2010 (UTC)

The "Virtualization for Security" book calls it "hardware virtualization", but only when it's hardware-assisted virtualization like amd-v (probably not understanding you still need a beefy amount of hypervisor software to complete the picture). Pcap ping 20:41, 21 April 2010 (UTC)

The "Linux in a nutshell" book call is "full system virtualization". Pcap ping 20:49, 21 April 2010 (UTC)

"The best damn sever virtualization book period" calls it "hardware virtualization". Pcap ping 20:51, 21 April 2010 (UTC)

Virtualization: from the desktop to the enterprise also calls it "hardware virtualization". Pcap ping 20:54, 21 April 2010 (UTC)

Some Fedora 10 book also calls it "hardware virtualization", but again seems to confuse it with hardware-assisted virtualization. Pcap ping 21:01, 21 April 2010 (UTC)

Another security book also calls it "hardware virtualization", but again referring to the hardware-assisted one. Pcap ping 21:03, 21 April 2010 (UTC)

Some "Principles of Information Systems" textbook also calls it "hardware virtualization". Pcap ping 21:06, 21 April 2010 (UTC)

Some Citrix/Xen exam prep book also calls it "hardware virtualization". Pcap ping 21:08, 21 April 2010 (UTC)

Some MBA-type book also calls it "hardware virtualization". Pcap ping 21:08, 21 April 2010 (UTC)

Mendel Rosenblum uses "hardware-level virtualization". Pcap ping 16:52, 22 April 2010 (UTC)

Summarizing, although there is much confusion among authors, "hardware virtualization" seems to win by fair margin. Pcap ping 21:12, 21 April 2010 (UTC)

Problem is it's too vague. As soon as you do this, people will then again add crap about virtualizing some hardware, like memory, ISA, peripherals. "System virtualization", consistent with Smith etc, expresses that it's the whole system that is being virtualized.

Nevertheless, if you manage to get the whole mess cleaned up, I might get involved again. heiser (talk) 19:06, 22 April 2010 (UTC)

Virtual Machine Facility/370 is not Hardware-assisted virtualization

Virtual Machine Facility/370 (VM/370) did not provide hardware assisted virtualization in 1972. It relied on normal address translation and trapping of privileged instructions in problem state.

The VM Assist (VMA) came later, and was only a very limited assist. It was only with the introduction of the Start Interpretive Execution (SIE) instruction (see IBM. 370/Extended Architecture/Interpretive Execution. SA22-7095. {{cite book}}: Unknown parameter |separator= ignored (help)) that VM/XA and later versions of VM truly provided hardware-assisted virtualization. Shmuel (Seymour J.) Metz Username:Chatul (talk) 19:14, 28 June 2010 (UTC)

Hardware Virtualization Disaster Recovery

I wanted to get your thoughts on a new section for hardware virtualization

A Disaster Recovery (DR) plan is good business practice for a hardware virtualization platform solution. DR of a virtualization environment can ensure high rate of availability during a wide range of situations that disrupt normal business operations. Continued operations of VMs is mission critical and a can compensate for concerns of hardware performance and maintenance requirements. A hardware virtualization DR environment will involve hardware and software protection solutions based on business continuity needs. Hardware virtualization DR methods

Tape backup for software data long-term archival needs

This common method can be used to store data offsite but can be a difficult and lengthy process to recover your data. Tape backup data is only as good as the latest copy stored. Tape backup methods will require a backup device and ongoing storage material.

Whole-file and application replication

The implementation of this method will require control software and storage capacity for application and data file storage replication typically on the same site. The data is replicated on a different disk partition or separate disk device and can be a scheduled activity for most servers and is implemented more for database-type applications.

Hardware and software redundancy

This solution provides the highest level of disaster recovery protection for a hardware virtualization solutions providing duplicate hardware and software replication in two distinct geographic areas. —Preceding unsigned comment added by Kwesterh (talk • contribs) 14:49, 5 November 2010 (UTC)

Is KVM paravirtualization?

The section "Paravirtualization" says "This system call to the hypervisor is called a "hypercall" in TRANGO, Xen and KVM". AFAIK KVM is not considered paravirtualization... am I right? If so 'KVM' should be removed from the above sentence. — Preceding unsigned comment added by 129.88.69.157 (talk) 16:26, 10 August 2011 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified 3 external links on Hardware virtualization. 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/20110516030950/http://www.visionsolutions.com/downloads/whitepapers/EssentialDR.pdf to http://www.visionsolutions.com/Downloads/Whitepapers/EssentialDR.pdf
• Added archive https://web.archive.org/web/20120815054111/http://www.drj.com/new2dr/w2_002.htm to http://www.drj.com/new2dr/w2_002.htm
• Added archive https://web.archive.org/web/20100923021435/http://www.vmware.com/files/pdf/DR_VMware_DoubleTake.pdf to http://www.vmware.com/files/pdf/DR_VMware_DoubleTake.pdf

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

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 04:49, 30 October 2017 (UTC)

Don't like lead section

I'm a little leery about the phrasing in the lead. "Virtualization hides the physical characteristics of a computing platform from the users, presenting instead an abstract computing platform" seems to refer more to software virtual machines used by programming languages, I.E the JVM or P-Code, instead of full virtualization which the main article seems to focus on, which does aim to emulate the physical characteristics of machines. Not removing it immediately, but would like thoughts on this. Sources supporting alternate definition:

https://www.cl.cam.ac.uk/research/srg/netos/papers/2003-xensosp.pdf - Xen and the Art of Virtualization - "In a traditional VMM the virtual hardware exposed is functionally identical to the underlying machine"

(Not OA) - https://dl.acm.org/doi/10.1145/1168917.1168860 - A Comparison of Software and Hardware Techniques for x86 Virtualization - "Software on the VMM executes identically to its execution on hardware, barring timing effects. ... An overwhelming majority of guest instructions are executed by the hardware without the intervention of the VMM."

I'd propose changing it to "Virtualization emulates the hardware environment of its host architecture, allowing multiple OSes to run unmodified and in isolation. In environments with hardware accelerated virtualization, the overwhelming majority of instructions are executed by the hardware," or something similar. Rcfische2 (talk) 19:01, 20 August 2024 (UTC)

In some ways an operating system could be looked at as "[hiding] the physical characteristics of a computing platform from the users, presenting instead an abstract computing platform", e.g. by supporting device-independent I/O. Hardware virtualization is more specific; your new lead better expresses that. Guy Harris (talk) 07:39, 13 September 2024 (UTC)