Linux TCO: Less Than Half The Cost of Windows
ggruschow writes "Linux Today reports 'The cost of running Linux is roughly 40% that of Microsoft Windows, and only 14% that of Sun Microsystem's Solaris, according to a new study which examined the actual costs of running various operating systems over three years.'"
They throw in a few eye catching facts, such as this:
The Windows technicians, however, only managed an average of 10 machines each, while Linux or Solaris admins can generally handle several times that.
Good enough for you?
The PDF for the study is hosted on IBM's website... I'd be willing to bet that it was IBM that commissioned the study. Anybody know?
;-)
**begin sarcasm**
What a big suprise that would be if a study funded by IBM finds that their Linux solutions perform better than Windows and Sun!
**end sarcasm**
That said, it is nice to have some pro-Linux FUD out there!
Ben
So the article says that Linux web servers are cheaper to run than Windows ones. Is anyone surprised by this? Is it actually news?
I'm still waiting for the article to come out discussing TCO as it relates to desktops, which is where most of the money is lost in support dollars....
I am assuming that the Linux and Solaris admins are using the CLI to manage the servers via SSH but I believe the slowest way to manage a server is through a keyboard and mouse -- pointing and clicking away. Most of the Windows servers I have managed in my career were through a GUI interface using a remote control program like PC Anywhere and Microsoft's Remote Admin software. With Linux, Solaris and now Mac OS X Server, I use SSH and a keyboard to do my work. With shell scripts and other tricks, I can blaze through server management that I would never be able to do in a GUI environment at the same speed. Even with Mac OS X Server's great GUI management tools, I prefer to fire up Terminal and remotely manage the system through a CLI -- or maybe I just long for the days of my Apple ][.
On the other hand, with the massive numbers of zombied Windows machines probing my networks, it could be that Windows-only Admins are just plain idiots with a MCSE which accounts for the productivity gains of Linux and Solaris admins.
Strange women lying in ponds distributing swords is no basis for a system of government.
Start a timer, tell each admin they need to perform some obscure task. Now see who gets it done first, assuming their skills are the same for their respective systems. A windows person might be lightening fast because they just have to click in a couple of places. Then again, they will probably have to reboot. A Linux person might need to check the info (or man) page and pray it is well written for their part then try to implement it. However they will most likely not have to restart anything except that particular service. Solaris? Well I suppose that depends on the version.
Last year we decided to migrate off Windows.
We first moved to OpenOffice. Painful, when your clients all use MS Office, but it's possible.
Now we're moving to Mozilla-based browsers.
All our servers (except one) went to Linux in the last year or two.
Now we're killing the last Windows desktops, putting Lindows-OS in their place.
Apart from the license savings, everything just runs better.
There is a huge fear of change, and this works in Windows' favor.
But there is no doubt that open software is better built and cheaper to run.
Changing costs something. But there is no doubt about the TCO of Linux (and its applications) being lower.
Sig for sale or rent. One previous user. Inquire within.
So, let me get this straight... they actually expect us to believe that WinAdmins make almost 70K a year to handle 10 machines? I don't know what kind of fantasy world this study was done in, but I want in!!!
Of course, far be it from me to suggest that this portrayal of WinAdmins might be a bit off... but, for reference, I support close to 200 WinTel machines and 5 servers, and I don't make anywhere close to 70K US a year... I think I may print this article and see if I can get a raise out of it...
Lesse... 70K per 10 boxes, 200 boxes, equals... woohoo!
If I knew the wedgies I gave you back in 6th grade would have resulted in this . . . I might have taken a moments pause.
If they want to do TCO of apples to apples, maybe they should have used Sun LX-50 x86 box, instead of their Enterprise class machines. There are feature sets in that class of machine (eg domaining, redundant hardware, hotswap etc) that are just not available in an x86 box. The cost of those features greatly inflates the TCO for Solaris.
"The greatest dangers to liberty lurk in insidious encroachment by men of zeal, well-meaning but without understanding."
At the last place I worked there was one *NIX admin (me) who was responsible for 15 HP-UX machines, a couple of Sun boxen, one lone AIX machine, and about 10 Linux boxen, I was also the webmaster and one of the NT/2K admins. There were 4 dedicated NT/2K admins for about 25 machines and they were always busy working on the machines. If it wasn't exchange dying or the DNS crapping out on us it was some wierd WINS issue or a virus. These guys were very talented admins too, not the MCSE tripe that comes out of the pipe now. Linux TCO is much much lower, I've been preaching this for a long time. I'm not saying that *NIX is perfect or never has a problem, but it is much rarer and the problems are usually easier to fix than the self corruption that goes along with Windows.
(B) + (D) + (B) + (D) = (K) + (&)
Half the price and half as useful!
God, poor karma.
Seriously though folks, there is a reason Windows is expensive as it is. It's part corporate greed, but it's also part quality of product.
I run Linux on quite a few web servers and a database server here and there and I love it for my servers. It's fast, stable and black and white. Just like I like em. I run Windows on my desktops. Recently I decided to install RedHat 7.3 just to see how things were coming on the desktop front. It was pretty smooth, until I decided I wanted to move my mouse, at which point the X installer locked up.
Seems the X mouse drivers can't figure out a plain ol' optical mouse running through a Linksys KVM. I searched and searched to no avail. The mouse just resets over and over, every second or so.
Screw it. Back to the server room with you! Where's my XP CD?
The idea that a Linux admin can handle more machines then a Windows admin doesn't wash with me
/. then :o)
Then you've obviously never adminned both (or you're just lousy at it).
I personally admin over two dozen Linux servers... If we needed to, I could easily double that.. or triple it (although I wouldn't have time to read
16 or so of the machines I admin are squid proxies.. spread out over several thousand square kilometers.. a month ago there was a vulnerability reported in squid (not too serious - only affected unsecured boxes), and it took me about 90 minutes to patch them all, including compiling the software and testing it on our dev machine to make sure that it worked in our config (which it didn't right off the bat - some of the directives in squid.conf had changed.)
Windows does take more to config. As Malor said in another reply, scripting is everything.
Where did they get those machine to tech ratios? Maybe small businesses that only have a few machines anyway. The shops I have worked in and friends in the industry the ratio is MUCH higher anywhere from 50 to 1 and up. We were mainly Windows shop and added Unix systems to the mix as time went on. In fact our management kept showing us reports from industry groups showing 100 to 1 as a common ratio. My argument was and still is what type of systems were they? In an ISP or ecommerse site with farms of servers the ratio can be high, because a system can go down with minimal effect. I worked more on large databases, and business systems and when systems were down it affected revenue and we had to get them back on-line ASAP.
Articles like this don't do Linux any good, when management see bogus numbers in them. This is not a an artice I would show to management to try and get Linux system integrated.
At UCSD, the Microsoft Assimila... err Microsoft Student Rep essentially crashed our "Linux Setup Day" event a couple years ago and handed out free copies of Windows 2000.
.net, etc. right now on campus.
The MS student rep would give out free copies of Windows, Visual Studio, Office, etc. on campus, and I'm sure that Microsoft is giving out free copies of WinXP, Visual Studio
The thing is that Microsoft did NOT do this sort of thing on such a wide scale until Linux grew in popularity on campus.
MS is obviously trying very hard to keep CS students from learning to seriously develop software outside of a Microsoft Environment.
By providing students with MS software for free, they hope to stop students from using open source development tools.
If Windows is all programmers know, that is all they'll develop for.
"You spoony bard!" -Tellah
I was principal architect for Excite Clubs for 4 years. During a period of one year, we went from 100K page views to over 20M page views a day.
,the page generation times went down to .1 seconds and the highest observable CPU load was less than 10%.
We had a rather unique situation. We started the project on Windows NT 4.0 and later migrated to Win2K. During that time, we were barely able to handle 1M page views per day on the windows boxes. In addition, the average page generation time was 2 seconds. The 20 windows boxes we had in production cost approximately 17K a piece (quad compaq proliant with 1 gig of ram) and were all experiencing 80% or more CPU usage.
The 20 boxes were managed by 1 sysadmin (6 years experience from MS consulting services) with a full time assistant. This does not count the high school students we had wandering the racks hard rebooting terminally ill boxes.
Most admin time was spent on upgrades, boxes that would just stop working (we called it spontaneous server rot) and trying to use a host of opaque, inadequate tools to detect and eliminate bottlenecks. Build, rollout and staging tools were also a big time synch. Finally, the installation of software onto a new machine in the right order with all configuration parameters took an extradinary amount of time.
In addition, I had one full time engineer writing noting but 'nanny' programs to monitor the program and restart the process when there were problems.
With all this work, the system still went down daily.
After much politicking we translated the program to JSP (straight page per page translation) and moved to solaris machines. The java middle tier ran as on solaris. The 20 compaq boxes were replaced with 16 solaris boxes. Oddly, we paid almost the same amount per box (20K versus 16K).
Immediately, we were able to more that 5M pageviews per day with no changes to the software. In addition
Our sysadmins were replaced with a part time (less than 5 hours per week) solaris admin. The roll out scritps were trivial to write and maintain. We had very few upgrades/security patches.
Most important, the host of tools provided to monitor system performance and tell exactly where bottle necks were and the truly deep understanding of the system internals by the sys admins allowed us to eliminate the remaining problems and scale to 20M pageviews per day.
That is right. two orders of magnitude better performance for precisely the same code. And and order of magnitude less admin time.
Those were measurable results. Here is my 'opinion' of why the differnces were so dramatic.
I taught Win32 programming and system internals for four years. I was also chief scientist for Redmond Communications who publish a technical journal on Microsoft Software/strategies. So I am not a linux bigot.
My observation has been, that no one truly understands the internals of a windows system. Just as I start to get a handle on the latest caching, memory management, threading issues, there is an 'upgrade' via some patch that changes many of the internals. In addition, as shown by the above threads, most windows sys admins seem to have vastly difference experience and understanding of how to configure and maintain systems.
Unlike most nerds, I will not blame the admin, but blame the system. In the scientific community, windows, in practice, has proven to be somewhat opaque.
Unix, on the other hand, is incredibly well documented and all source is available. Uncertain how inodes are locked and released? No problem, there are many books and references to help you. If worst comes to worst, crack open the damn code.
This has nothing to do with open source, but more to do with the which communities evolved the techonlogy and the underlying motivations of companies hawking their wares.
Note, this is not a good thing, or a bad thing it is only a thing.
There were many people out there criticizing the studies accuracy. I must say I do not have a single colleague that I have spoken with that doubts its varisity from personal experience on BOTH sides of the isle. I just knew that I had to share my own experience with you. My only doubt about the story is that I would say 'order of magnitude' for production servers.
Thank you for your time,
Carmine Mangione
Even though I'm not a heavy scripter my story follows:
I used NT/IIS 4.0 for several years switching to 2000/IIS 5.0 when it was available. I have a small business and primarily use my websites for testing solutions that are implemented for my clients and for e-mail. That being said I had to check my servers daily for hacks and patches and got rooted several times. After switching (sorry Apple) to Linux I've been rooted 1 time (my fault for leaving a known bug open via ftp). Going from checking daily (sometimes 3 to 4 times a day) and still getting hacked, to checking weekly (unless I notice an article here a la openSSL, etc.). My TCO is dramatically less. It has also allowed me to confidently recommend Linux solutions at my full time job.
Time is $$$ and the less I spend trying to avoid script kiddies the more time I have to do real work and get paid.
Scripting is close to the true reason linux admin takes less effort.
With linux the admin's education is never limited because everything is open to inspection. With windows the admin's education is limited to what M$ wants them to know. Thus severly handicapping their diagnostic abilities and their intuition.
This effectively reduces the effect of eXPerience for the windows admin. A 5 year windows vet will likely be no smarter than a 3 year vet. However, a 5 year linux vet has every opportunity to exceed his 3 year counterpart.
Computers are getting fast enough that we don't need firm-ware intregration of everything and can apply some abstraction layers/components. If we abstract the following features:
1. Execution environment (like a run-time engine)
2. Database
3. File System (use #2 instead?)
4. Networking protocols
5. Workspace managers (desktops)
6. Graphics and hardware interface
behind standard paradigm-neutrual protocols, then the God-Damned OS does NOT mean diddly squat.
I have too much existing Windows software to just chuck Windows. I don't want to depend on MS, but I don't want to start over. Thus, if you want to make MS irrelavent, then make the OS irrelavent using/making the above standards, then we don't have to marry neither Bill Gates NOR a smiling penguin.
F the OS wars. Think beyond it people. Think abstraction and standards. Windows will shrink when standards make it so that you don't need Windows, not because Linux crashes slightly less.
The Penguin can go fuck Clippey for all I care. You are all fighting the wrong war.
Table-ized A.I.
There's just one thing I have issues with:
MS products might be 20% more likely to be infected than Linux just based on the rollout numbers, but experience repeatedly shows that MS products in the real world are infected much more often than that.
Now, that 20% difference you speak of may be limited only to exploitation of bugs, but to limit your comparison only to that is the same sort of mistake you accuse the authors of the paper of. You can't just limit yourself to exploitation of bugs, you have to include exploitation of design flaws as well, and that is where Microsoft products typically fall on their face. Whether it's automatic execution of malicious code during a document preview in the explorer (or Outlook) or an install program that doesn't make you change the sa SQL server password, Microsoft has consistently shown that basic security is something of an afterthought to it. If they're changing that, then good for them! It's about time.
But until they start actually designing their products with as much consideration of security as of useability, Linux will maintain a significant advantage over Windows in resistance to attacks.
Use 'slashdot stuff' in the subject line in any email you send me if you want to get past the spam filter.