Gartner: Linux Servers Booming

Tarantolato writes "According to a recent Gartner report, low-end Linux server shipments grew significantly in the first quarter of 2004. Part of this may be due to the comeback of the relational database market in 2003, where Linux growth was especially strong, while Windows growth was weaker. There is mixed news for Sun, who saw growing shipments but declining revenues in Q1 of 2004."

Comments

Meh, statistics

[stratjakt]

To say linux server sales are up 27% means little if the volume is low.

If I sold one last year, and three this year then I can talk about 300% growth, but that number is meaningless.

Yeah, linux is gaining ground, but has a long way to go.

Re:Meh, statistics

[BiggerIsBetter]

I wonder what sort of increase the MacOS server market showed? It also means little if servers are being shipped with XP or even no OS, and being loaded with Debian after delivery. I doubt it takes into account systems built in-house either. Statistics show only what gatherer wants them to show.

Re:Meh, statistics

[jbplou]

What moron would by a system to be a database that comes with XP lets be realistic if you are shopping for server hardware you can buy it without an OS and also the MS OS would be Windows Server 2003 or 2000 possibly. Database servers aren't run on desktop pcs.

Re:Meh, statistics

[Unknown Relic]

It also means little if servers are being shipped with XP or even no OS, and being loaded with Debian after delivery.

I know in the past when ordering servers from Dell even though we order them with no operating system preinstalled the sales rep would ask what OS we were going to be using, presumably to gather just that sort of information. As for stripping windows and installing a Linux distribution, how often does this really happen on server hardware? On desktops, sure, but on a server? It's highly unlikely any serious hardware could even be ordered with a non-server version of windows, and if you're footing the bill for that, chances are it's not so you can just toss it and do a reformat as soon as the machine arrives.

Re:Meh, statistics

[nelsonal]

$900 million in quarterly sales is not exactly low volume. I'm not sure how much of the revenue is mainframe based (they also had a good quarter due to a hardware refresh by IBM. Unit numbers are not disclosed in the press releases (you gotta pay for the details) but unit growth was just behind revenue growth, which is backward from other markets where prices usually decline, they've been ramping for several quarters in linux servers. I think windows servers (or perhaps the whole x86 market is about $4 billion/quarter.

Re:Meh, statistics

[macfreak12]

all our 1u HP servers come w/o an OS loaded on them and are not licensed for windows.

Re:Meh, statistics

[hdparm]

RTFA. Says there, 1.57 million units sold.

Re:Meh, statistics

[Vengeful weenie]

You really haven't seen much if you think that all data is housed only on enterprise class servers. I often have the distinct impression that most data is stored on old dekstop machines that never get backed up, and rarely get patched.

Re:Meh, statistics

[Anonymous Coward]

I took delivery 2 weeks ago of a high end server that I stripped of Windows. The vendor had installed Windows to test the box coz thats all they new how to install.

Re:Meh, statistics

[RoLi]

And that gets modded up as insightful?

Depending on whose numbers you are going to believe, Linux already holds about 30% to 50% of the market, strangely the Linux share is always higher in areas where the numbers are not guessed but counted like in webservers where Apache/Linux holds a comfortable majority.

Have you ever searched a webhoster in Germany that even offers Windows? Mine stopped to offer it last year. Windows is dying there, and losing more and more:

look here

In Japan, the same picture:

stats

In a lot of countries, Windows on servers is already an exotic niche platform.

Webhosters don't want it anymore because the support costs aren't worth it and the added risk (a worm was the reason my webhoster stopped offering Windows) has to be paid somehow. Customers don't want it anymore because Apache gives them a much larger palette of availabe webhosters - thus more choice, lower costs and more competition among webhosters.

Windows just offers no real advantages to make up for all the license hassles.

Re:Meh, statistics

[sforman]

200% growth, I think you'll find.

Re:Meh, statistics

[topdogqqq]

Statistics can be misleading, but I know at our University many of us no longer order servers with windows. Just this week we are ordering 3 servers with no OS that will get Debian installed. When I look around I see many others going this way too. It's not at critical mass yet but it is substantial and I see the writing on the wall, Microsoft will be slammed hard. They won't die by any means but they will be humbled to the level they deserve within a few short years. Every server I buy without a Microsoft product helps and I have no doubt that they feel it. Just their FUD based marketing against linux shows that they are already being impacted. It's real hard for a server salesman to sell MS Windows servers when the customer says, "But, I can get Linux for free and it's better". Rock on

Re:Meh, statistics

[Lonewolf666]

quote: Have you ever searched a webhoster in Germany that even offers Windows? Mine stopped to offer it last year.
I live in Germany. In my favourite computer magazine C't there are frequently ads of webhosters who offer servers with both kinds of OS for rent.
Usually, the Linux root server is slightly cheaper than the equivalent Windows server. A typical price would be 49 euros/month for a small linux server with limited transfer volume, and 59 euros/month for the windows version. Now what does this tell you about the TCO? ;-)

Re:Meh, statistics

[Anonymous Coward]

That initial acquisition cost is swamped by the ongoing maintenance costs in just a couple of years. I would still buy Windows servers if they were more reliable, but they aren't! By the same token, I am not switching our servers to Linux because it is free; I am switching them because Windows is just a PITA to keep running and secure!

Re:Meh, statistics

[Anonymous Coward]

> As for stripping windows and installing a Linux distribution, how often does this really happen on server hardware?

As often as M$ forces unwanted costly upgrades.

DISCLAIMER: just my personal opinion.

Re:Meh, statistics

[Mateito]

> but that number is meaningless.

You don't work in Marketting, do you :)

Re:Meh, statistics

[solferino]

Thanks for posting those links. While I appreciate how much more significant the move to linux is outside the U.S. those graphics still blew my mind. Seeing a pie chart showing 90% to an OSS solution with microsoft at 7% and dismally trailing out into oblivion makes me delirious with joy.

Unfortunately the country I live in (au) is not quite at this joyous point yet. Oh well, maybe once we disengage from the U.S.'s latest military empire play we can start thinking a bit more independently. I live in hope.

Re:Meh, statistics

[jbplou]

The article was about new computer sales for databases, I really don't think companies were buying old desktop machines and being included in this article.

Re:Meh, statistics

[matuscak]

Interesting links. What I think is fascinating is the percentages for IIS seem to be highest in .gov (37.86%) and .mil (66.13% !!). At least from march to april, the IIS percentages for those domains *grew*.

Re:Meh, statistics

[mwood]

Obviously not -- the folks in Marketing know that "that number is meaningless" and "that number is not something we can use to stomp the competition" are completely unrelated concepts.

Great news, but is there a typo?

[coupland]

According to Gartner, revenue of Linux-based server hardware rose 57.3 percent over the first quarter, while commercial Unix server revenue fell 2.3 percent.

Is it just me or does 57.3 percent growth genuinely impress you as well? I can only assume the article contains a mistake since it claims 57.3 percent revenue growth for linux-based servers over the first quarter which means "in three months". This strikes me as unlikely, unless Linux is actually destroying everything in its path. Shouldn't this read year over year in which case the 57.3 percent growth happened in 12 months, not 3. Can anyone confirm for sure? Regardless this is fantastic news, it's been a many, many years since we've seen genuine competition in the OS market.

Re:Great news, but is there a typo?

[0racle]

Percent increase isn't impressive without hard numbers. Were there 100 or 50,000 shipments of linux servers the year before? Shipping an extra 57 this year isn't much, but an extra 28,650 is. So try not to wet yourself over this, there are no actual numbers, just some nice sounding percentages.

Re:Great news, but is there a typo?

[j0hndoe]

Regardless this is fantastic news, it's been a many, many years since we've seen genuine competition in the OS market.

Really? I've been reading articles for YEARS about how Linux was Microsoft's biggest threat. Starting with those insane "Linux IPOs" in 1999/2000. And Microsoft pointed at Linux in the anti-trust trial. (ok, that was self-serving, but it ended up being true anyway). This isn't really anything new at this point. The numbers are just confirming it.

Re:Great news, but is there a typo?

[AndroidCat]

I looked at the Gartner site, but the main page and a search didn't turn up this report. Is it available to the public yet, and does anyone have a link? It's usually best to go to the source to check the numbers to see what they really mean.

Re:Great news, but is there a typo?

[Anonymous Coward]

And they buried those hard numbers quite deep in the article. First paragraph, and it sounds like quite a few:

According to data released this week by research firm Gartner (gartner.com), worldwide server sales were up 27.1 percent in the first quarter of 2004, reaching 1.57 million units.

Re:Great news, but is there a typo?

[flight666]

RTFM. Linux _grew_ by $180million dollars compared to $100million for MS. Those look like actual numbers to me.

Re:Great news, but is there a typo?

[Galvatron]

I can only assume the article contains a mistake since it claims 57.3 percent revenue growth for linux-based servers over the first quarter which means "in three months"

When they say things like this, they usually mean "relative to the first quarter of last year." So if all four quarters show a 50% rate of growth, the growth rate for the year would be 50%, not over 400% (1.5^4-1). They do things this way because the season can make a big difference in puchases, and so they don't want to muddle things by comparing different months or different quarters. The same thing happens for big tetail chains (Walgreens generally reports sales growth on the order of 14% each month, but they mean relative to the same month last year, not the month immediately prior).

Re:Great news, but is there a typo?

[Anonymous Coward]

Gartner doesn't generally give away their reports. Business buy them, and they aren't cheap.

Re:Great news, but is there a typo?

[PhrostyMcByte]

Please direct me to this manual of yours. I would like to grow my funds by $180 million also.

Re:Great news, but is there a typo?

[BasilBrush]

TO to gartner, click on Media relations and the press release is there. The actual report will cost you money though.

Re:Great news, but is there a typo?

[rozz]

RTFM. Linux _grew_ by $180million dollars compared to $100million for MS. Those look like actual numbers to me.

didn't u foget some numbers? like:

"Windows server platform ... new license revenue reaching $2.8 billion"

"New license revenue for Linux RDBMS totaled $299.3 million"

an ant and a double-sized-ant compare quite the same with an elephant!

Re:Great news, but is there a typo?

[sumdumass]

so it's to quick to say "it looks like the year of linux"?

Re:Great news, but is there a typo?

[juan2074]

You forgot that extra .3, which is the partial server my company bought.

Oddly enough, .3 dollars is how much the Linux installation discs cost (about 10 cents per disc).

Unfortunately, that is 50% more than the company spent on the installation media for Solaris 8 for the Intel platform.

Re:Great news, but is there a typo?

[0racle]

$180 spread across how many companies compared to $100 million for one means MS is moving more product then any one Linux vendor. I bet Unix sales also were over $100 million but everyone still says Unix is going the way of the dodo. So still these numbers are nothing special.

Re:Great news, but is there a typo?

[jelle]

They are something special, if the elephant is growing because it is between farts and the ant is growing because it's working out like crazy. Sure, right now it looks like an ant sitting down there, but a rocket launch starts from the ground as well, but ends way up going 17500 mph or more in space.

Keep the noise down!

[Anonymous Coward]

Linux servers are booming, Windows servers are bombing and crashing. It's hard to get caught up on my sleep!

Re:Keep the noise down!

[Anonymous Coward]

I work for the IT dept. of a large academic institution, and about 70% of our servers run some form of Linux, while around 10% run Windows Server. Guess which ones have the best uptime.

Re:Keep the noise down!

[Spua7]

Yippy this weekend is reinstall our Win2K DC time.

Oracle versus SQL Server

[ergo98]

This isn't Linux versus Windows -- it's SQL Server versus Oracle. Shops are choosing Oracle and then choosing Linux as the platform (given that it's largely irrelevant what platform it runs on). The submission implies that it was a toss up between Windows and Linux, and after choosing Linux they started looking around for a RDBMS.

Re:Oracle versus SQL Server

[sloanster]

it's largely irrelevant what platform it runs on

um, Oracle would beg to differ with you on that one - and speaking for myself as a customer, I am keenly interested in what platform it runs on...

Re:Oracle versus SQL Server

[flinxmeister]

I wish my company would do this. They chose Oracle on Linux a couple years back with a single (mediocre at best) developer as the admin. The resulting and expected crappy experience resulted in a decision to go with Sun.

But instead of going with serious Sun hardware we now have a whole slew of entry level sun servers that routinely sweat under even moderate loads. Even then, they just incrementally buy small servers for *every* function. (It's like NT shops used to be, except twice as expensive).

The server admins have, at this point, kind of a glazed humorous look in their eyes. If we'd made an intelligent choice between Linux and Solaris (even if it resulted in a hybrid situation), we could have saved thousands and thousands of dollars and have a managable situation. Instead, it's insanity.

Word to the wise: Don't believe the hype and TEST THE SETUP WITH COMPETENT ENGINEERS!. You'd think this would be common sense...but alas...we seem to be made to suffer. It's our lot in life.

Re:Oracle versus SQL Server

[iabervon]

It doesn't matter to most customers. Most customers don't plan to interact much with the OS on their database servers, and want to use whatever Oracle says will be best. They just want the best-supported configuration, which means that they'll use exactly what Oracle installs (or has Red Hat install), and won't change anything. Of course, it's a significant milestone therefore that Oracle is pushing Linux (and moving to it internally), because that means that it actually has the best TCO for that kind of system, because Oracle knows that they get all the money that doesn't go into the customer's TCO on the database.

Re:Oracle versus SQL Server

[frfriel]

Do I see a nitch market here? We have an enterprise Oracle database running on Linux (On an IBM eServer in VM). Very cool setup, and it doesn't break a sweat. It is administrered by two admins and supported by numerous programmers and DBA's. What I am saying is if I weren't a government agency with deep pockets, I might want this kind of setup on an outsourced basis. Hey, looks like kaching kaching to me.

Re:Oracle versus SQL Server

[1001011010110101]

Actually, Oracle is OS agnostic. They run everywhere (bah, mostly. Win, Linux, Hp Ux, Solaris, and a big etcetera).
They do pitch linux a lot lately.

Re:Oracle versus SQL Server

[sapbasisnerd]

Can you please clarify, "On an IBM eServer in VM" could be any of half a dozen diffent things. Do you actually mean Oracle on Linux on eServer zSeries running as a guest OS under VM? Which Linux? SuSe?

Re:Oracle versus SQL Server

[Anonymous Coward]

Man! They should have listened to you. Maybe they could have gotten YOU to take time out from your busy day of posting on /. even.

Why didn't they check with you? They could have did IT right. It's just sad.

Re:Oracle versus SQL Server

[kfg]

Oracle has little interst in the TCO of its customers, except insofar as it can promote it to its customers.

Oracle is interested in Oracle. It is not in Oracle's interests to have its fortunes tied entirely to single propriatary OS whose owners can dictate Oracle's business to Oracle.

There is, of course, a cautionary tale to be found here by Oracle's customers as well, and TCO is not the be all and end all of the matter. It is often worth paying more to achieve some desirable end, say, independence from a single monolithic supplier.

And if Microsoft's products were truely and clearly superior you can be sure that instead of touting a lower TCO they'd be perfectly happy to tout the fact that they're a bit more expensive, but worth it.

In fact, when the whole fallacy of their TCO argument blows up in their face this rather the tack I expect they will shift to.

They may find, however, that the time has finally come when they must come about and run before the wind rather than beating into it.

Which brings us right back to my original premise.

KFG

Re:Oracle versus SQL Server

[j0e_average]

Of course they would pitch it...when comparing license costs for Oracle vs SQL Server, the latter wins hands downs...but considering that in order to use SQL Server, one must also have a license to 2000 Server, the price difference is less of an issue for Oracle if linux server is employed.

Re:Oracle versus SQL Server

[sapbasisnerd]

I don't think it's so much Oracle vs. SQL Server as it is a multiplatform DBMS vs. a uniplatform one.

With Oracle and DB2 and several open source options lined up on one side and SQL Server on the other.

Still today most applications are limited by the scaling of a single node database and every time a company with MSSQL realizes they are running out of gas and and facing a choice between an expensive and risky conversion to another platform or an expensive and risky rehost onto Windows Datacenter they don't make the "one platform" mistake again (or one hopes, in reality of course some PHB heavy shops do exactly this over and over again).

I also think that the day most businesses actually really care about which DBMS they use is somewhat past. Databases are becoming more commoditized and more and more people are buying applications (or application suites, or application server platforms) first and then asking what's the "normal" database to run this with.

As long as MS and ORCL continue to make products that are directly competetive with the likes of SAP and Peoplesoft the answer these companies are increasingly going to make is "DB2 or Open Source".

Re:Oracle versus SQL Server

[frfriel]

I am sorry. It is running as a guest OS under VM on an eServer zSeries and it is SuSe. Very intuitive sap.

Re:Oracle versus SQL Server

[stokkie]

The costs of licensing is not such a big an issue. The cost of ownership the more. (the huge big ass Oracle databases excluded)

Re:Oracle versus SQL Server

[fritz1968]

Most customers don't plan to interact much with the OS on their database servers, and want to use whatever Oracle says will be best.

I believe that there is another reason that Oracle is "recommending" the linux platform and I am a little surprised that no one has yet to metion it: Oracle CEO Larry Ellison.

I thought that it was common knowledge that Ellison hates MS. Or at the very least, he wants his company to be better than MS. If that is the case, then why would Oracle want to "recommend" that their customers use the Windows platform to run oracle? That is just more money into MS's pockets. By "recommending" linux, then there is less money going to Microsoft.

Another thought is that by keeping MS out of the picture and recommending Linux, the less likely that their client will be stolen away by MS in favor of the MS SQL server. I wonder how many times (if any) that Oracle lost a client to MS SQL server because the consultant who setup the server suggested SQL Server?

just a thought...

Re:Oracle versus SQL Server

[Mr. Piddle]

(given that it's largely irrelevant what platform it runs on)

Apparently, you never experienced the joys of Oracle 8.1.5 on Windows NT. Oracle really is at home on UNIX/VMS; Windows is just window dressing for Oracle.

Re:Oracle versus SQL Server

[Mr. Piddle]

One thing I just thought of is that by eliminating OS licensing costs, Oracle salespeople can make a pitch for selling just a little more of Oracle's products. "Since you saved X by not buying Windows...how about adding in Oracle Extension Pack Y?" Most companies will simply choose to save the money, but any company choosing to shovel it over to Oracle is good for Oracle.

Re:Oracle versus SQL Server

[sumdumass]

Well doesn't oracle have thier own operating systems based on unix/linux? I believe it is basically set up for a 64 bit system though. Also using linux as a substitute instead of windows would let then concentrate more on sales instead of development and open an enviroment that was previously not a real player. The enviroment is smaller companies not wanting to invest in "big iron" platforms. I know oracle has for a while supported windows but I believe the performance and other aspects of thier product shine on unix like systems.

So in adition to your comments, taking the "we support linux" line they have effectivly opened themselves up to a hole new line of business opertutities. And with such new opertunities, they don't have to worry about microsoft purposely breaking an aplication with an upgrade (like serving a different web pages to different browser in an attemp to make it look inferior). As I see it, microsoft Now has to compete with the stability of linux on thier server market that was running oracle on windows. This not only provides a benefit for thier consumers, it Stops the threat of waisting profits on suport calles after an update broke the aplication. This was a GOOD MOVE that has seen increased benifits.

Re:Oracle versus SQL Server

[Anonymous Coward]

But when comparing features and reliability, the latter still has a long way to go.

Re:Oracle versus SQL Server

[CatOne]

Oracle is keenly interested in the TCO of its customers.

Specifically, because given a fixed TCO, if the OS is "less expensive" there is more margin for Oracle in the sale.

For example, 5 years ago, a customer would buy Oracle to run a "big database" on an E10K with Solaris. They may pay $1M for a 3 year contract for the Oracle license, and $1M for the hardware and software maintenance from Sun (these numbers are rough, order of magnitude figures).

These days, the customers can run the Oracle database on cheap, commodity Linux. And use Oracles "RAC" functionality to load balance the DB across the boxes. So the hardware "piece of the pie" changes to $5K (decent dual CPU Opteron with 8 MB of RAM), times 15, so $75K for hardware. They pay Oracle $1.2M for software, RAC support, and Oracle supports their RHEL 3.0 installation (and kicks back some money to Red Hat).

So NOW, Oracle's piece of the pie is better. AND EVEN MORE, the customer is now even more tied in to Oracle's DBMS, because they're using the 10g RAC functionality for load balancing, fault tolerance, failover, scalability, yadda yadda. What happens in 3 years? Yep, license costs go up.

But again, overall TCO has improved in this case... DBMS costs have gone from $2M every 3 years, to $1.5M... a 25% savings. And the sales rep and the buyer can go golf together at Pebble, because they both saved/made more money for their companies.

And Sun can lay off more people, because their bigwig fat cat sales managers can't realize that the last E10K was sold about 18 months ago, and their margins are going straight to /dev/null.

Re:Oracle versus SQL Server

[iabervon]

Oracle actually provides almost exactly that (outsourced databases). They all run Linux, according to the sales rep I talked to a year ago.

Re:Oracle versus SQL Server

[leandrod]

> how many times (if any) that Oracle lost a client to MS SQL server because the consultant who setup the server suggested SQL Server?

Problem is, when running on GNU/Linux I'd recommend IBM DB2 or PostgreSQL, both higher-quality, cheaper, more ISO SQL compliant than Oracle.

Mod Parent Down

[Anonymous Coward]

He speaketh against our penguin overlord. Without Tux what will we worship? CowboyNeals nutsack?

Re:Mod Parent Down

[SnoBall]

Did anyone forget about the BSD Daemon? I mean, come on, BSD is sometimes used in server environments... so at least give it a 15-minute plug

Re:Mod Parent Down

[Fjornir]

Uhm. Someone chose to be called "SnoBall", and asks for a "15-minute plug"...

Sorry, dude. Back to the 18 1/2 minute gap for you.

Re:Mod Parent Down

[Anonymous Coward]

The two are not mutually exclusive, and would be most interesting if surgically joined.

What was Microsoft's server growth?

[Anonymous Coward]

Just for a point of comparison, I'd like to know. Linux grew 57 percent, Unix was down 2 percent, what about MS? (I'd just like concrete confirmation of Linux kicking Microsoft's fanny, OK?)

Re:What was Microsoft's server growth?

[Anonymous Coward]

No, I don't think so. Mickeysoft stuff has always sukked like a hoover. Anyone with half a brain got rid of anything called 'mickeysoft' a long time ago. Any lingering sentiment by mickeysoft fanbois shows dillusion venturing on madness. Kick the heavy drug habbit. Sober up and face reality. Rebooting from failures costs money. Even scheduled shutdown because of instability costs money. Mickeysoft has shown itself to be craptacularly bad for years and years. Read mickeysoft's SEC filings. They are peeing their pants scared because they know Linux is beating them like a rented mule. Deal with it!

Re:What was Microsoft's server growth?

[Canberra Bob]

If you had read the article, Linux isnt kicking MS in the slightest. MS sales were UP, and their market share stayed the same.

Linux market share increased, but certainly not at the expense of MS.

Re:What was Microsoft's server growth?

[Anonymous Coward]

"Anyone with half a brain got rid of anything called 'mickeysoft' a long time ago."

When they got rid of 'mickeysoft' they must have bought Microsoft then, because they sure didn't put in linux.

Is because competitors are not free market

[argoff]

Copyrights have nothing to do with free market property rights, but are rather like government regulations about what people can do with information. But the GPL, has found a 'loophole' in these restrictions - and is far more accountable to free market forces. People who have closed software are going to continue to pay huge opportunity costs as the market takes off again.

Re:Is because competitors are not free market

[Anonymous Coward]

What loophole? The GPL is just another license, and it depends on existing copyright law for its validity.

The holder of a copyright can allow you to make copies of their if they so choose. They might, if they desire, require you to do something in exchange for that right. A licensor could, for example, require you to stand on your head each Tuesday morning while rubbing blue mud in your navel in order to have the right to copy their work. Most licenses simply ask for money. The GPL happens to ask that you release the source and changes to the source (among a few other things.)

There's nothing particularly "loophole-ish" about these requirements. They're all of the same form: "If you do X, you can copy my work". The very same principle that gives closed-source authors the right to charge for their work is the one that gives GPL authors the right to their requirements. If there's really no such thing as intellectual property, "information wants to be free", and copyright are all bogus, then so is the GPL. Fortunately, that's not the case.

Re:Is because competitors are not free market

[Anonymous Coward]

Think of the GPL as fighting fire with fire. Just bacause you use it, doen't mean it's good to burn down forests.

not consistant

[networkGhettoWhore]

I am not trying to undermine Gartner, but this poll seems to be inconsistant with the recent netDeck poll which stated linux hardware rose 31% as opposed to the stated 57% here.

Re:not consistant

[Anonymous Coward]

Well, often enough Gartner comes out with a survey that happens to be pro-Microsoft, in which case Slashdot is inundated with posts about how Gartner is just a bunch of corporate shills that don't know their asterisk from their octothorpe.

In which case, you might as well disregard these results as well. They're no more reliable than the others.

Re:not consistant

[Jim Hall]

I am not trying to undermine Gartner, but this poll seems to be inconsistant with the recent netDeck poll which stated linux hardware rose 31% as opposed to the stated 57% here.

Keep in mind that Gartner tends to survey a different set of customers/companies than netDeck does. So it's probably fair to say that among the largest companies (the Gartner survey) Linux hardware rose 57%, but more industry-wide (the netDeck survey) it only rose 31%.

Deja Vu

[__aanonl8035]

I swear every quarter I hear this same news story.

From 2001
http://librenix.com/?inode=984

The report shows Linux server revenue rising from 2,422,266,299 in 2001 to 9,142,634,360 in 2005 and total units rising from 543,778 to 2,610,235 over the same period.

End-user research done in 2000 presents a good picture of the real market share of Linux as a server operating system and serves to project the probable market share for Linux this year, as well as a Linux server forecast through 2005.

but...

[tisme]

But... those Windows Server advertisements say that Windows server is so much faster... And the results are from INDEPENDENT and ACADEMIC sources?? Why would companies buy the inferior product?? Surely Microsoft would not lie? :P

Re:but...

[Canberra Bob]

Pity that Windows sales increased and their market share remained constant.

Dont let the facts get in the way of a good joke.

Re:but...

[Anonymous Coward]

How did they manage that trick? If sales increased but market share did not, that would imply that Linux is selling twice as fast as Windows is.

Re:but...

[Canberra Bob]

All you need is for the market itself to grow. If the market is growing, and your market share remains the same, it means you are selling more units, hence your sales increase.

Windows sold roughly $3 billion in RDBMS systems, Linux systems accounted for roughly $300 million. If Windows sales increased by 50%, and Linux sales increased by 100% what position would they be in? Windows would be selling $4.5 billion (increase of $1.5 billion), Linux would be selling $600 million (increase of $300 million). Windows market share would be blowing Linux away even though Linux sales were increasing at double the rate of Windows. Market share is not only based on how many you sell, but also how many total units were sold by everyone.

Now if Linux sales were increasing a larger amount in dollar terms to Windows, and Windows sales were dropping in dollar terms, then you can say that Linux is taking MS market share. Until then, just because Linux grows at a great rate means nothing. As one of the parents pointed out, if I sell 1 unit this year and 3 next year thats an increase of 300%. Does that mean that Windows market share has somehow dropped because of my massive increase in sales? Of course not. Same principle.

Re:but...

[rozz]

Why would companies buy the inferior product??

did u ever hear about the price-quality ratio?

Job Market

[pavera]

I have definately seen this in the job market
1 year ago I was looking for DBA jobs, and hardly anyone requested linux knowledge/experience.

Now I'm looking again and I would say for 70-80% of the jobs I look at (DBA stuff) linux is either recommended or required. Linux really is making alot of inroads into the DB server market from what I see.

Re:Job Market

[markan18]

I created myself a linux job by selling linux and openbsd solutions to one of my customers. Linux do a great job replacing aging and worm ridden NT servers.

Furthermore, when you search linux on yahoo hotjobs , you will find thousands of linux jobs. In Québec, were i live, linux jobs are also available.

Linux Servers Booming?!

[Pan T. Hose]

Oh my God, I'll better watch out for my server, I don't want it to suddenly boom! I'll better check the water cooling system...

Re:Linux Servers Booming?!

[Anonymous Coward]

"Where's my earth-shattering kaboom?"

revenue vs other statistics

[xtermin8]

I was a little confused about the statistics in the article. Does this mean companies are just getting better at getting money for linux? How does this compare to the number of servers without linux sold? Any estimates of servers with other operating systems sold? I'm afraid there's just not enough context for the news to be meaningful.

In Related News...

[k4_pacific]

A Microsoft representative made the following statement at a press conference today:

"Where are you going? Come back! You'll all be doomed, DOOMED if you use Linux. DOOMED! After all, our studies, err, I mean independent studies have shown that Windows has a lower overall total cost of ownership. I mean, c'mon, Longhorn's coming soon. It will be better, we promise. It has Pallad-- err, Trusted Computing. Doesn't that sound nice? Trust? Can you trust Linux? You can? Fine! Be that way. We have FIFTY BILLION DOLLARS. We can by and sell your ass. Hmmppphhh!"

A followup press release attributed the remarks to an overly tight necktie.

Re:In Related News...

[Anonymous Coward]

What worries me is that you actually think you are funny.

Re:In Related News...

[Anonymous Coward]

And in another news conference today, a Linux GNU/Zealot made the following statement:

"This is the year of Linux on the desktop! It is, really it is! And if you use M$ products you will burn in hell, burn I say! Gnu/Linux is the roxor!!!"

These remarks were attributed to a condition known as Open Sores, believed to be due to the large swarms of flies accompanying zealots from the Linux temple.

Gartner

[Anonymous Coward]

So do we like Gartner today?

Re:Gartner

[NonSequor]

Timeo Danaos et dona ferentis
(I do not trust the Greeks, even bearing gifts)
--Vergil, The Aeneid

Re:Gartner

[Anonymous Coward]

Yes (0.8 likelihood)

Re:Gartner

[Stevyn]

Nope, tomorrow they'll come out with a report saying that IT spending has gone up 50 times due to migration over to linux. However, this is hedged by microsoft's TCO being -$88,000 per server. Yes people, it actually generates money on the fly!

Re:Gartner

[femto]

We like Gartner on Mondays, Thursdays and Saturdays of odd numbered months. In even numbered months, the rule is Tuesdays, Fridays and Sundays, with the exception of the first week when Sundays become Wednesdays. In leap years, swap the role of odd and even months.

Consequently, yes, today we like Gartner.

Of course that depends on what time zone you like in. If you live in the US, you should still be hating Gartner until midnight.

Re:Gartner

[killjoe]

Who is we kimo sabe?

Re:Gartner

[Too Much Noise]

That would be a compliment Gartner doesn't deserve. The Greeks were feared for being subtle. Gartner is only following the money.

SCOIX

[Anonymous Coward]

Well this is bad news for innovation and copyright. Linux on the rise == Piracy on the rise. This is a sad day for us all. I hope this "rise" of Linux servers means people are buying the correct LICENSE from SCO to run Linux. Otherwise, we're all in trouble. SCO has given the world so much, when they invented Linux, and now nobody wants to pay for it and pretend its free. Nothing is free folks. Get on over to SCO and buy a license and sleep better tonight.

You're all a bunch of smelly ass hippies, and have no business using a computer if you don't want to pay for the OS!

Re:SCOIX

[dcstimm]

thanks for letting me know, i didnt know the slashdot croud felt that way, I happily payed for all my linux licenses from sco. Only thing I didnt like was the fact that they charged me almost double for my dual cpu machines. Oh well better be safe than sorry. Hopefully they will lower the price alittle in the future so linux will become more mainstream.

zeitgeist

[Coneasfast]

nice to see it's growing, zeitgeist still shows a pitiful 1% though :(

Re:zeitgeist

[Anonymous Coward]

I think the parent makes a valid point, double figure percentage increases in the server market mean little when your OS counts for a mere fraction of the desktop market. You may try and modbomb this away, but in the end Slashdot is not the world, googles zeitgeist however, is.
Linux still has a mountain or seven to climb

Re:zeitgeist

[1lus10n]

thats desktop, not server. More importantly that number cannot be accurate since just about every method of checking that can be modified. (and in many cases it is)

Re:zeitgeist

[Anonymous Coward]

Except that, since about half of the poorly designed websites on the planet stupidly check for IE or Netscape for windows, there could be many people who use other systems but are forced to masquerade as something else. When they get complaints about how their site works fine with x (for all x=opera, omniweb, safari, konqueror, etc) pretending it's IE, they just try and make the ID process more stringent.
In a word, I don't trust zeigeist's numbers.

Re:zeitgeist

[Anonymous Coward]

I can't seem to read the chart there (it's quite small). Where did you find the 1% number?

Re:zeitgeist

[mini me]

Linux has more desktops than MacOS, and MacOS is at 4%. So if you take Linux + Other you get 4%, which might be pretty accurate, though I think it's more like 3%.

Re:zeitgeist

[ry0n]

I have 5 computers running (or almost running) at the moment. Only two of them ever visit websites / use google. Oh, and I'm not a business. All of these machines are for my entertainment and/or utility. If I understand the zeitgeist correctly, those computers don't exist. Or, more simply - who surfs the web on a headless database server?

Re:zeitgeist

[Canberra Bob]

Blasphemy!

2004 is the year of Linux on the desktop! It is, seriously!

Only 7 months and Linux will dominate, honest, really and truly! We promise!

Re:zeitgeist

[Gheesh]

nice to see it's growing, zeitgeist [google.com] still shows a pitiful 1% though

On the other hand, the Netcraft Web Server Survey shows 67% of the machines running Apache, and most of them run Linux or FreeBSD

Re:zeitgeist

[Anonymous Coward]

I think the parent makes a valid point, double figure percentage increases in the server market mean little when your OS counts for a mere fraction of the desktop market.

Actually, they mean a decent amount. Only not in the desktop space. Personally I'm fine with MS continuing its desktop monopoly for a little while longer as long as infrastructure continues to become Free.

Revenue vs numbers

[xtermin8]

The important number for me would be units sold. The revenue increase could be that hardware companies are getting more money for linux pre-installed. The "traditional" method has been to install distros after the hardware has been bought (perhaps with no OS?)

More Servers with Less Money Spent

[yintercept]

Skipping the Linux v. Windows v. Sun debates. The main gist of the article is that there are more servers being thrown up at a significantly less cost.

To a very large extent, this is just the gradual realization of productivity increases. The scary side of the equation is the extent to which companies are pushing people out of the equations. The ever dropping margins means a tougher job market for slashdotters. Or, how should I say it. More work for lower pay.

Word to the wise: Don't believe the hype and TEST THE SETUP WITH COMPETENT ENGINEERS!

The expectation of lower costs leads to scenarios like the one described where the company is trying to get by on one subpar admin, or they push their support staff to the brink with more servers than the staff can handle...without a good plan for installing or using the servers.

Re:More Servers with Less Money Spent

[flinxmeister]

That's what's odd in this case. Once the decision was made to go with Sun they actually hired a couple very good Sun guys, and a couple really good Oracle DBAs. But the momentum for lots of low end Sun boxes was already in full swing by the time they loaded the new red Swingline and had enough clout to matter.

I guess the danger of many cheap servers is that it can give the illusion of cost savings over the long haul, when in reality a far more economical approach would be to go ahead and start big...or just scale as you go (which is supposed to be a strong point of Sun). But it's alot easier to periodically squeak through POs for 10 low end servers than one high end server at half total price.

And the true costs of juggling those servers go beyond dollars to many, many hidden things. Unlike stories at Google and archive.org, I don't think most companies are firing up a beowulf cluster to run accounting apps.

I know that some companies skimp on talent in line with lower hardware costs, but this is the second time I've seen this illusion in effect. The first was with the aforementioned NT environment where a few good but overwhelmed engineers were left to try and keep a hopelessly large NT farm afloat...while the execs wondered why they couldn't see Emperor Uptimes new clothes.

Re:More Servers with Less Money Spent

[Keeper]

Skipping the Linux v. Windows v. Sun debates. The main gist of the article is that there are more servers being thrown up at a significantly less cost.

You've obviously never seen how much it costs to purchase an Oracle license...

Re:More Servers with Less Money Spent

[bubbha]

The scary side of the equation is the extent to which companies are pushing people out of the equations. The ever dropping margins means a tougher job market for slashdotters. Or, how should I say it. More work for lower pay.

But does not the parent say that his company ended uo NOT saving the money they could have? So the proplem is that they don't know HOW to save money. That implies that they don't really know what they are doing and how much it's costing them. How can Linux help that?

MIcrosoft already knew it.

[Stonent1]

When I worked there, there were posters saying that if each MS employee converted 5 linux servers to 5 windows servers that MS could finally outsell linux in the server market. I SO wanted to take a picture of it but I didn't want to get caught.

Linux Servers Booming

[frfriel]

The bottom line is this. The number of servers sold with Linux preinstalled is increasing. The sales of Linux built for multiprocessing is increasing. But, is it increasing enough to become a true competitor in the market. To say that sales are up 57% by revenue is mileading. Especially if revenue previously was crap. I could say my income increased 600% if I got a raise to about 12,000 a month. But there are tons of people who make 12,000 a month. Linux sales don't even scratch the big guys (or guy). If the revenue (and/or # of servers shipped with Linux) continues to increase at a 57% clip, then we will soon be seeing some drama in the market. May the penguins day come, and it's sun shine bright enough to blind the other guy.

Re:Linux Servers Booming

[ckaminski]

I'm sorry? Two years ago the effort to get ANY server from Dell or HP with Linux preinstalled wasn't worth it. Now they're shipping in volume, and you're asking if it's increasing enough to be a true competitor...

Doh... have I just fed a troll? ;-)

Re:Linux Servers Booming

[Anonymous Coward]

"Linux sales don't even scratch the big guys (or guy)"

I think the $2.8 billion of Linux server sales last year, or the projected $4 billion of Linux server sales this year represent quite a big scratch.

It took Windows a decade to grow from a niche $100 million of workgroup servers to their present $15 billion or so per year. That's starting the count from the point where analysts noticed Windows NT, not the day Bill hired Dave Cutler. I'd be very pleased to see $10 billion sales of Linux servers in 2007, about a decade after analysts first noticed Linux server sales. It's a tougher market now than when Microsoft arrived, and it might be tougher to get those all-important migrations, but to reach $10 billion Linux has to take away Windows sales in serious volume.

What about MS?

[sharkey]

The Northwind database will be encrypted in 2005.

They are taking Linux seriously

[Anonymous Coward]

I know Red Hat is not linux, but it is to be noted that after the MSCI rebalance they included Red Hat in Prime Market 750

What is the total number of units shipped?

[IAmMaxHarris]

If the volumes we're talking about are low, it doesn't say much to say that "Linux sales for servers are up 27%"!

Someone who sold a single orange last summer can turn around this year and sell six oranges to his friends and say, "Wow! My market share went up 600%!". However, this means very little.

Without real numbers, all one can correctly say is that Linux is gaining market share. And judging from everyday experience, it has a long road ahead of it.

Re:What is the total number of units shipped?

[Anonymous Coward]

Numbers of units really are better than dollar sales. You really can outfit 4 or 5 Linux servers for the cost of outfitting 1 Microsoft server. Every thousand dollars RedHat makes means a loss to Microsoft of about one million, so dollars to dollars comparisons mean little. A few years ago, Linux servers had about 12% of the market. If they are up 57%, they should now have about 19% (or if you do the math, 18.84%). Most pundits will say 'see, Microsoft is still winning, nothing has changed'. But, if the 57% increase happens next year too, and the year after that, then Linux will have 46.43% of the server market. One year after that, 72.9. Just more numbers to throw about.

Re:What is the total number of units shipped?

[slmdmd]

it is total revenue divided by revenue per unit.

Re:What is the total number of units shipped?

[matt4077]

Yeah, and 3 more years after that, Linux will be at 115,56% market share.

What is a low end linux server?

[cft_128]

I wonder what they qualify as a 'Low End Server'? Only uniprocessor? Quad Xeon with an ultra320 hardware RAID? Any x86 Linux box?

Re:What is a low end linux server?

[jelle]

The terms "Low End" or "High End" for lawyers and accountants like the Gartner people are has nothing to do with technical specs and everything with price point. If that 1 Ghz PIII loaded with software costs a bundle, it's high-end. If you bought 10 GHZ of CPU power for $2k loaded with software, it's low-end.

%'s from the Article

[WaterBottle]

It looks like the total Linux based DB market of $300M was just slightly bigger than the increase in MS based market (3% of 7.1b = $222m) Big percentage changes, but different market shares to start...

"sales increased" not "market share went up"

[IAmMaxHarris]

Doh! I didn't mean "market share went up". The correct phrase was "sales increased". That's what I get for posting so hastily...

Much better write-up of same data

[ToLu the Happy Furby]

There's a much more detailed summary of the Gartner report up at com.com. The overall numbers are thus:

Total WW Q1 server revenue: $11.81 billion, +9.3% quarter-on-quarter*

That breaks down into:
Windows: $4.13 billion, +19.5%
Proprietary Unix: $4.02 billion, -2%
Mainframe: $1.7 billion, +12%
Linux: $1.02 billion, +57.3%

That leaves $.94 billion unaccounted for; I was thinking this chunk could be VMS and NSK revenues, but that makes it difficult to fit HP's 32.5% share of x86 revenues into the $.94 billion left over when you subtract it plus HP's $1.17 billion in proprietary Unix sales from HP's $3.07 billion total sales. (And that's ignoring HP's Q1 IA64 sales, which were very substantial.)

Of course all these questions are surely answered in the report itself, but I'm not gonna pay 95 bucks to find out.

*How do I know the figures in the com.com article are QoQ and not YoY? Because the Gartner summary (linked above) puts overall YoY revenue growth at 24.1%, not the 9.3% reported in the article. Which makes both the 57.3% Linux growth and the 12.5% Sun decline even more stunning.

Re:Much better write-up of same data

[Torulf]

OK, let's do some forecasting. What every slashdotter wants to know is, of course, when Linux will be bigger than Windows. Based on the numbers in the articles and linear forecasting, here's what we get*:

Servers
Q1/04 Q2/04 Q3/04 Q4/04 Q1/05 Q2/05 Q3/05 Q4/05
Win 4,0 | 4,8 | 5,7 | 6,9 | 8,3 | 9,9 | 11,9 14,3
Lin 1,0 | 1,5 | 2,2 | 3,4 | 5,1 | 7,6 | 11,4 17,1

Databases
2002| 2003| 2004| 2005| 2006
Win 2,7 | 2,8 | 2,9 | 3,0 | 3,1
Lin 0,1 | 0,3 | 0,8 | 2,0 | 5,2

So, basically we can come to the conclusion that Linux will surpass Windows as a server platform by the end of next year, both on platform sales and on database sales.

There, now I told you what you wanted to hear, so mod me up!

*No, this is not realiable. It is extremely unlikely that this quater's growth will continue for 2 years

Re:Much better write-up of same data

[martin-boundary]

That leaves $.94 billion unaccounted for;

Well yes that's right. They're using a Pentium.

Re:Much better write-up of same data

[Frit Mock]

Not realy ... the growth in one quarter is typicaly compared to the same quarter the year before in financal reports.

To make a forecast, Linux will be at 1.5 in Q1/05 and at 2,2 in Q1/06 ... an then let's see what influence Longhorn will have on the market.

Re:Much better write-up of same data

[kwoff]

You have to do it in gigadollars.

Re:Much better write-up of same data

[sapbasisnerd]

I wouldn't assume these questions are necessarily answered in the full report.

In any case the math does work, if we make the not unreasonable assumption that for both Windows and Linux IA64 adoption is similar and that it is low enough to be in the noise we get the following.

((4.13+1.02)*.325)+1.17 = 2.84 leaving HP with .23B of the unaccounted for stuff which would be the remaining dribs of VMS and Tandem stuff. I think it's safe to assume that a large chunk of the remaining .71B is IBM iSeries nee AS/400 and that there are some other minor issues like a few remaining dribs of proprietary file and print NOS and so on.

A much more interesting question is does Gartner double count zLinux and Linux and Mainframe and how are they going to account going forward for Linux vs. proprietary UNIX when customers can buy a box that will a) run both and b) decide themselves how to slice up the box's capacity and can change it at any time...

Re:Much better write-up of same data

[ToLu the Happy Furby]

I think it's safe to assume that a large chunk of the remaining .71B is IBM iSeries nee AS/400 and that there are some other minor issues like a few remaining dribs of proprietary file and print NOS and so on.

Ah--OS/400 (or whatever they call it these days). That's what I was missing.

With that in place the numbers make much more sense.

Re:Much better write-up of same data

[sapbasisnerd]

OS/400 (or whatever they call it these days) Well for a few more weeks it's till called OS/400, when the next release (V5R3) ships they will change the name to i5OS.

Re:Much better write-up of same data

[jak163]

How do I know the figures in the com.com article are QoQ and not YoY? Because the Gartner summary (linked above) puts overall YoY revenue growth at 24.1%, not the 9.3% reported in the article.

Original article says Dell revenue up 24.8 percent for year earlier period not overall market--unless you are referring to a different link.

From the article: "Gartner's data showed that Dell is the fastest-growing of the leading server vendors, increasing unit sales by 38.4 percent and revenue by 24.8 percent over the period."

Those increases are year over year as is standard when reporting quarterly growth. Quarter to next quarter is referred to as "sequential" growth.

Re:Much better write-up of same data

[Chemicalscum]

The big question is what was the sales of servers sold without an operating system installed. We know Dell an the other hardware manufacturers sell them to you and we know that they will all have Linux or a BSD installed on them (OK a few might have Solaris or yuk SCO put on them) and none will have windows installed.

This could make a big difference to the overall picture.

Re:Much better write-up of same data

[Anonymous Coward]

ToLu the Happy Furby (63586) wrote: How do I know the figures in the com.com article are QoQ and not YoY? Because the Gartner summary (linked above) puts overall YoY revenue growth at 24.1%, not the 9.3% reported in the article.

I hate to have to do this, but the percentage numbers are for different things. The revenue grew at 9.3% (to $11.8 billion), and the units grew at 24.1% (to 1.57 million boxen). These numbers are from InfoWorld. The article even said that it is "a clear move by users towards low-end servers and the Linux operating system"; that is a reasonable explanation how lots more boxes can ship with a smaller increase in cost.

FWIW, an article from SmartMoney.com from today uses an IDC report, and got similar (but not identical) numbers and it stated that the comparasons were Year-over-Year (total sales: $11.5, unit growth: 22.4%, Linux sales: $900 million, Linux $ growth: 56.9%, Linux unit growth: 46.4%).

Re:Much better write-up of same data

[InformationOverload]

From what I've seen, a linear forcast might be too optimistic because of the change in Redhat licensing fees. Companies and universities now have to pay an annual per machine "support" fee to use a RedHat distribution. The cost of deploying Linux on hundreds or thousands of machines just went up dramatically for many of the existing and future Linux install bases. I'm curious how others here have reacted to this change. Have you modified your Linux strategy? It's probably just a hiccup in the relentless march towards Linux dominance, but it seems to me like another distribution will need to pick up the slack in the Enterprise server space before installs continue at the historical rate.

what a coincidence

[Anonymous Coward]

just today i saw an advertisement saying something like, "get the facts, Microsoft Server 2003 is faster than Apache on RHE"...of all places, here on slashdot. anyone have a link to those claims? i'm not in the market, so i mostly ignored it ;)

Re:what a coincidence

[ckaminski]

Sure it is. If you're comparing .Net applications to CGI-bins... it could easily outperform Apache on RHEL. Then again, Apache+PHP would crush IIS w/ CGI-bin or ASP for that matter...

Now, Apache+PHP vs. IIS+PHP... that's a contest I haven't rigged up yet.

Re:what a coincidence

[fwarren]

I have been up for 20 hours and feel too lazy to google it out for you.

In a nutshell, take a box, put Windows 2003 on it, have some engineers from Microsoft come out, tweak the box, filesystem, registry, services etc for optimal performance serving static web pages, then blast those pages out to an intranet.

Then take a box, put RHE on it, get an 11 page document on optimizing linux for serving web pages, and follow all of the instructions on page 1 only. Leave services running, do NOT tune the file system.

In a nutshell, Microsoft paid to have a 3rd party run a web server under the conditions where Microsoft would perform the best and Linux would perform the worst, and then tune the web servers even further to make sure that would be the case.

It would be like us having a race, both of us with the same make and model of car, and I hire a professional driver, and have the car tuned up by a professional pit crew, and you take yours to jiffy lube and have your grandmother drive. Would you be surprised that my car outperformed yours?

---------

Re:what a coincidence

[Anonymous Coward]

ASP .Net + IIS would crush a LAMP solution just because ASP .Net it tightly woven into the OS and doesn't have to worry marshalling data between Apache, PHP, and the Linux Kernel in a losse knit way.

In 2002, 2 Windows Server: 1 PAID Linux Server

[Anonymous Coward]

From IDC's 2002 stats from last fall, there was 1 (23.1%) PAID Linux server for every 2 (55.1%)Windows Servers sold.
http://news.com.com/2100-7344_3-5088233.htm l.
This does not include the Linux servers created from free downloads.

With the massive increase of Linux servers, what is the ratio between Windows Server against PAID Linux server.

I have 2 servers that were stripped of Windows.

[khasim]

They were initially purchased to run Windows apps. When the Windows servers were upgraded, I grabbed these two and put Linux on them.

So, four sales for Windows (two initial servers and the two replacement servers)
-and-
No sales for Linux
-but-
Actual deployment is 2 servers for Windows and 2 for Linux.

(That isn't 50% of our servers. We have almost 20 Windows servers because the apps don't play well with each other.) I expect there are a lot more installations like mine out there. The sales percentages (particularly the $$$) will not tell you the real picture.

Not to mention

[bonch]

Linux is destined to replace those old UNIX systems companies are running. What's the surprise there?

When Linux rises above 1% usage on Google Zeitgeist, then it's time to stop the presses.

Re:Not to mention

[jayminer]

Linus himself is aimed at the desktop, he does not care about business, he leaves that to other people.

Re:Wow, those are some hard numbers!

[Anonymous Coward]

hahaha ASP loser.

The problem with that....

[Xtifr]

Of course, you can't replace five linux servers with five windows servers. If you tried, you'd have...not enough servers. You'd need at least ten to twenty-five windows servers to replace five linux servers. :)

Now, if you replaced five linux servers with five windows servers and four linux servers, it would look good for MS, which would apparently have more servers at that point, even though the linux servers would be doing 80% of the work. :)

Re:The problem with that....

[Anonymous Coward]

Why was this modded funny?

Here's an article that implies almost precisely that!

http://www.anandtech.com/IT/showdoc.html?i=1527

Cheap Linux Servers - Gentoo Linux

[the_oak]

I use linux for my servers as well, but it took me almost a day to find cheep hosting. And i could not even find what I was actualy looking for: Gentoo Linux Servers and a low pricing. I found a Debian server, 30$/month, very nice. And I installed Gentoo on it, cross-atlantic (a bit scary rebooting the first time :). If you use serverpronto and want Gentoo on your debian server there, I will be happy to send you the details of this adventure.

This must be FUD

[Anonymous Coward]

What? If it was a story about Microsoft someone would've said it by now

The problem with Gartner reports and Linux

[Glamdrlng]

Sometime last year I set up a linux snort server at work. Certainly a low-end server, by most standards. Did Gartner take this into account? I certainly didn't tell them, and I doubt they monitored me as I downloaded the iso's.

If gartner's stats are strictly based on data from redhat, IBM, etc, how can they possibly account for all of the "other" installs? I certainly hope these stats won't be used to calculate market share...

Wrong.

[britneys 9th husband]

If I sold one last year, and three this year then I can talk about 300% growth, but that number is meaningless.

What makes that number particularly meaningless is that that would actually be 200% growth.

Mod parent up, informative!

[sloanster]

I created myself a linux job by selling linux and openbsd solutions to one of my customers. Linux do a great job replacing aging and worm ridden NT servers.

That's an excellent business strategy, and I can tell you that there's a lot of business to be had just picking the low hanging fruit as you pointed out.

I've done a good deal of work for a local finance company which has been retiring ms windows servers and replacing them with suse linux. They are extremely happy with the improved reliability and performance.

Uh

[Anonymous Coward]

When Linux rises above 1% usage on Google Zeitgeist, then it's time to stop the presses.

Yeah, 'cause everybody knows that because those numbers represent accesses to Google that it corresponds exactly with Linux usage in general.
</sarcasm>

Changing attitudes

[Anonymous Coward]

At my university where I work in the IT department, if you would suggest a Windows server for anything today they'd have your head examined. Linux has become the defacto web and database server around here. It's not just that it's cheaper, it's better and you don't have to manage licenses which is a huge deal in a place like this where we can barely keep track of all the new machines that are constantly coming in.

I'm also very happy to see that when we place the order for personal computers for post graduate students, about 1 in 5 actually specifically requests a Linux workstation these days. That would have been unheard of just a few years ago.

Re:Changing attitudes

[Anonymous Coward]

Forgot to mention, I wonder just how many new Linux server deployments there really is. Our severs for example are just regular Dell server hardware with everything from P3's to quad Xeon machines. They all come without an OS installed, so they can't be included in any sale statistics.

Re:Changing attitudes

[16K Ram Pack]

I've been working with some younger guys, and they are very switched on to Linux. They also seem to be the kind of people who get the job done (whether on Linux or MS).

The movement of those people into the business world is going to have a dramatic effect on Microsoft's business.

Analysts run in packs - IDC numbers out tonight

[Anonymous Coward]

According to an article at Tekrati Industry Analyst Reporter tonight, IDC is saying Linux Server sales grew revenue at 56.9% and unit shipments at 46.4%. Also, they have a report stating Linux servers near the $1 billion mark in quarterly revenue. You can get to the IDC release off the Tekrati article or go to IDC directly.

Which Distribution, then?

[puntloos]

I'm sure this will cause fighting but I'm at the treshold of installing a new system with linux. And my question is: what's the best distribution to use for a production server? Im part of a tiny company so this is small scale of course..

As I see it there are 3 distributions eligible here:

- Debian - I've used this before in multiple servers, I like the dpkg system, though even the 'testing' distro set is a bit dated.

- SUSE - New distro out (9.1 I think?) and I've heard that this one is a contender, though I know little about it. There are some issues with hmm XFS or JFS I think.

- Gentoo - we're migrating away from freeBSD which was just not up to date and functional enough for our needs, so the portage system sounds an interesting option.

Re:Which Distribution, then?

[ewn]

- Gentoo - we're migrating away from freeBSD which was just not up to date and functional enough for our needs, so the portage system sounds an interesting option.

And on the first day, the Server said: emerge light!

Re:Which Distribution, then?

[BenjyD]

Six days to emerge the system? Sounds about right.

Servers bought with linux - used with pirated OS?

[Anonymous Coward]

..so.. How many of the servers bought without OS or with a "free" OS gets a pirated operating system installed, then?

Oooouch, karma-killer question.. :-)

You mock the sacred Slashdot ritual?

[Anonymous Coward]

You made the mistake of mocking the yearly cry that "this is the year of Linux on the desktop". You see, it is a sacred ritual here on Slashdot. Every year it is performed without fail. And every year, once that statement is made, it is never to be brought up until the following year.

So it has been, so it always will be.

All disbelievers who dare to challenge it will be modded into oblivion by the sacred zealot fanbois of the GNU/temple.

One huge problem with counting

[codepunk]

I don't know about you guys but I have two huge HP boxes that do a ton of stuff. These two clustered servers serve up 150 desktops, handle the company email, order entry processing, mysql database, postgres database, plone server, internal intranet, file serving.

Try doing that much stuff with two windows boxes. A windows installation rarely runs more than a single application.

What you really need to ask is what is the potential of those linux boxes that are shipped

Re:One huge problem with counting

[SpaceCadetTrav]

Not a problem on a pair of correctly configured clustered Windows machines. Can you explain why you think otherwise?

Faster to what? BSOD?

[Anonymous Coward]

just today i saw an advertisement saying something like, "get the facts, Microsoft Server 2003 is faster than Apache on RHE"...of all places, here on slashdot. anyone have a link to those claims? i'm not in the market, so i mostly ignored it ;)

Faster to do what? Context is everything.

If you believe Microsoft's marketing hype,

1. I've got a bridge in Brooklyn for sale, cheap,

2. There's this huge pile of money tied up in a Nigerian bank account that we can work together to free up, but it will require some small up-front costs...

gartner research

[rozz]

since when is Gartner research such a /. darling? i remember people demolishing Gartner (and research in general) just few weeks ago now they come up with some pro-linux numbers and suddenly they are THE research authority on /. ... bleah

SCO-1, RHEL+1

[SlashDread]

As of today, a Dutch Progress db software develop firm, is SCO-1, and RHEL+1, the other old crap will be fased out.

Yay for us! ;-)

"/Dread"

Windows was 69.4% of all servers shipped (Gartner)

[Anonymous Coward]

On a shipment basis, Windows dominated all others with 69.4 percent of the OS server market.

http://www.internetnews.com/ent-news/article.php/3 359981

No booming at our NOC

[jbrownc1]

Our smoke from time to time, but they never boom.

So does most Unix

[Anonymous Coward]

DUH

Hogwash

[Anonymous Coward]

Linux is mostly selling on cheap ass ~$5,000 dollar boxes anway, hardly powerful systems. Sure some are on big iron but that is a tiny percentage, Linux's claim to fame is commodity boxes and that is it's primary market.

Re:Hogwash

[jelle]

Cheap ass $5k per box? Where you do piss you boss's money away? It's not 1998 anymore when HP could catch $50k for four times more power than a $2k PII with Linux.

With Linux, that $5k box can easily saturate your 10Gbit LAN backbone. Actually, per individual box $2k tops is probably 80% of the market. Even tungsten (see below) is built with boxes that go for around $2k, but they probably spent more than that per box on the interconnect.

On the big and powerful side, Linux's claim to fame is its significant and growing presence list of biggest irons in the world, note number four and five running Linux and Thunder's speed and estimated ranking as an indication of what the next list will look like. Now that is worth some fame.

Laughable

[Anonymous Coward]

Windows is 70% of the market, *nix combined isn't even threatening it.

http://www.internetnews.com/ent-news/article.php /3 359981

Re:Laughable

[Anonymous Coward]

An AC wrote: Windows is 70% of the market, *nix combined isn't even threatening it. http://www.internetnews.com/ent-news/article.php/3 359981

But if you look at the sales numbers (which are posted in the same article), Windows is only 35% of the market. In fact mainframe sales (which can run Linux) were $1.7 billion (which is 14% of the %11.8 billion total market).

I do not doubt that Windows is dominating the low-end of the market, but Linux can go even lower and it can play on the biggest iron (something Windows simply can not do) being sold, giving it a wider possible market. That makes Linux a threat indeed.

Also note that the US only accounted for 37.8 percent of global server revenue. If we give the same amount to Europe (as a "worst case" scenario), that leaves 24.4% of the global market to places (like India, China, and South America) where the ratio of salary cost to license cost is radically different, leading to radically different TCO numbers. Nearly a quarter of the servers sales were in places where hiring twice the number of admins/developers an running Linux rather than paying the MicroSoft tax makes economic sense. That means that, over the long haul, Linux will win.

Windows Pirates Not Counted Either

[Anonymous Coward]

Pirated copies of Windows exceeds the number of non-commercial linux in use, especially worldwide.

Pirated copies of Windows not counted either

[Anonymous Coward]

Gartner doesn't count all the pirated copies of Windows either, which worldwide is estimated to equal 50% of all installs, or other words twice Microsoft's sales figures.

Re:Pirated copies of Windows not counted either

[Anonymous Coward]

However most of the 'pirated' copies of Windows are run by home users. Few businesses will risk running on pirated copies, especially with the threat of the BSA raids (less overseas of course). Those 'pirated' copies of Windows are also generally installed on machines that came originally with Windows, usually an older or lower-end version, probably, but machines that would still have been counted as a Windows sale (albiet perhaps a Windows desktop sale instead of a Windows server sale). I seriously doubt that a significant number of Windows server users are building their own boxes or buying machines without OS and loading unlicensed copies of Windows on them. Most of the people I know that build their own boxes or buy no-OS boxes run Linux or *BSD on them. The few home built boxes that I know of that run Windows are basically only used in a home environment.

While unlicensed Windows use may be huge worldwide, I seriously doubt it really has a significant impact on servers compared to the number of Linux and *BSD installs that are not a result of commercial distribution sales or OEM preloads.

A large percentage of Linux server deployments, even in a business environment are still machines that were OEM preloaded with Windows. That skews sales based figures in Microsoft's favor because every such instance doubles MS's market share relative to Linux for that sale. Even though progress has been made, it is still not as easy as it should be to get every configuration of server shipped with Linux preloaded or no OS loaded from many hardware vendors. A lot of them only have a few models that can be ordered without Windows, some none at all. And it is also common practice for people to re-purpose old machines that originally shipped with Windows to be Linux servers. I know personally of several boxes like that where I work.

Tinkering

[Anonymous Coward]

Linux requires lots of tinkering, no wonder students are interested. When they grow up they'll wan't somthing they can turn on, click a few buttons and forget, like Win2K3.

Re:Tinkering

[Anonymous Coward]

Hah. My experience and observation is that Linux servers are both easier to set up and less trouble to keep running smoothly than any version of Windows. If you think you can just click a few buttons on a W2K3 server and just forget it, then it will be owned or worm-ridden before long.

The days where Linux requires lots of tinkering have been over for a long time. I can set up a Linux database or web server in a small amount of time, and keeping it patched up is not a big effort compared to Windows server.

Wave height

[katorga]

Wave height is based on wind speed, duration and fetch. The linux wave is getting higher and higher.

My estimate is rought 18-20 months before the linux wave overcomes MS.

Spin-friendly title

[mwood]

Why do I get the feeling that I'll soon see one of those "get the facts" banners claiming that Gartner has determined installing Linux will cause your server to explode?

Re:Servers bought with linux - used with pirated O

[Anonymous Coward]

I personally doubt its very many. How many companies will risk running a pirated OS? If they get raided by the BSA they are screwed. Home users are a different story, but it is much, much harder to buy a home PC without Windows and few home users are adventurous enough to build their own, even though it isn't difficult these days. Most of those that do that I know of run Linux or BSD.

The "appliance" factor

[Steve Embalmer]

A couple of ago we chose Linux + PostgreSQL backend to power our health IS app suite. It's been a huge hit (> 400 units sold thusfar). We bundle service with the hardware, which is a simple configuration (no keyboard/display, etc)... as an economic decision it was great for our company. Our products *used* to run on SCO (before there was a Linux), resulting in bids with a ~$8K load for third-party software. Of course, oldSCO saw that money, we didn't. Linux was one the best things to happen to our invoicing.

I'm just wondering if increased Linux server sales aren't due at least in part to some of the same solutions being offered in vertical markets like healthcare?

Re:Wow, those are some hard numbers!

[Anonymous Coward]

ASP sucks donkey balls, it's a loser Web API for clueless ITT fucktards who should be flipping bugers instead of "coding".