Windows Update posts 100% uptime, beats Apple, Ubuntu

Royal Pingdom has posted the results of a short study that looked at the availability of the software update services for three popular operating systems: Windows, Mac OS X, and Ubuntu.
Pingdom's uptime monitoring service performed a test once every five minutes and if downtime was found, it was confirmed from two different locations and was counted for however long it lasted. The results were then tabulated in a graph:

Microsoft won with 100 percent availability (0 minutes of downtime), Apple came in second with 99.9 percent availability (2 hours and 34 minutes of downtime), and Ubuntu came in last with 98.64 percent availability (1 day, 5 hours and 45 minutes of downtime). Microsoft doesn't have much to be proud of, as it's the biggest of the three, and it's no surprise it took first place. Apple's 99.9 percent is also quite respectable.

Report a problem with article
Previous Story

Using Origami as a free alternate to Windows Media Center

Next Story

HP TouchSmart PC Released

80 Comments

Commenting is disabled on this article.

Has anyone thought that maybe Ubuntu's update servers (or mirrors) had the biggest downtime because 8.04 was released during this time period?

Just saying. They do have to mirror almost seven hundred megabytes of data to thousands of downloaders. Similar situation happened back during the Vista beta 2 days when MS's servers nearly crashed or slowed to a crawl over thousands of people downloading a 2+ GB ISO.

This is just fanboy food, another excuse to argue. These statistics don't help consumers like us at all.

(RPDL said @ #29)
This is just fanboy food, another excuse to argue. These statistics don't help consumers like us at all.

But this is the truth. It's a just comparison between the update experiences from 3 popular operating systems. What are they supposed to do, lie so that Ubuntu or Mac gains marketshare unjustly? Or maybe they should lie so that Windows fanboys don't once again say that 'oh look, time and time again windows proves to be the best and most reliable!'.....Please.....

Microsoft only has the highest marketshare because it has the highest third party software support and they have already established themselves in most corporations and homes around the world. It's just that they got people's trust first, it doesn't mean they are better or worse. It's just the way it is.

Who really cares? I never had any problem getting fixes on any operating system. A good practice is to cache updates anyway so you don't have to tax your WAN... I'm surprised to see them acknowledge Linux though, interesting.

Microsoft - 100% FACT

It's funny how Apple fan boys always find ways to distort the facts and make them in their favor... And it's the same with OS X, Iphone, Apple computers and so on. And Intel used to suck, now it's the best... bla bla bla

ARGH

Distorting what?

That we have the best OS? The best (virus free) user experience? The best interface? The best phone? The best design?

And Intel used to suck, absolutely. Enter the AMD Athlon series - Thunderbird, Thoroughbred, Palomino, Barton, etc., that used to spank competing Intel cpus consistently. Now Intel has finally put some R&D and $$$ into its products, and made a very wise choice by hooking their wagon to Apple's star.

There is no distortion. The only real, consistent success story in the industry for the past few years has been Apple. They make products that people want - exciting, interesting, easy to use products. It's pretty simple.

(LTD said @ #24.1)
Distorting what?

That we have the best OS? The best (virus free) user experience? The best interface? The best phone? The best design?

And Intel used to suck, absolutely. Enter the AMD Athlon series - Thunderbird, Thoroughbred, Palomino, Barton, etc., that used to spank competing Intel cpus consistently. Now Intel has finally put some R&D and $$$ into its products, and made a very wise choice by hooking their wagon to Apple's star.

There is no distortion. The only real, consistent success story in the industry for the past few years has been Apple. They make products that people want - exciting, interesting, easy to use products. It's pretty simple.

WOW You are so blind.
The best OS is Windows has it has the biggest user share.
The best user experience is also on Windows because more users experience it and enjoy it together
The best interface is N/A as both are great and different users like different things
The best phone is also the most outdated phone to come out.
The best design is also sketchy as I personally dont like the all white.

Intel has ALWAYS been the upperhand in the (home) processor department. AMD has never (except for one flawed year which IMO Intel actually let themselves slip because the monopoly police would have been all over them) beat Intel.

Its pretty simple that you can keep lying to yourself and others. But whatever floats your boat.

Oh great here we go again. People it's called an opinion, everyone has one and they're all different. Get over yourselves and your stupid "My OS is best" arguments.

(powerneowin said @ #24.2)
...
WOW You are so blind.
The best OS is Windows has it has the biggest user share.
...
Yup. And McDonalds has the best burgers, because they sell the most burgers!

Most sold = best.

(markjensen said @ #24.4)
Yup. And McDonalds has the best burgers, because they sell the most burgers!

Most sold = best.


I know you're using sarcasm, but according to the majority, that's exactly what it means. They're the best to the majority of people. If they weren't, why would they sell the most?

(Eis said @ #24.5)
I know you're using sarcasm, but according to the majority, that's exactly what it means. They're the best to the majority of people. If they weren't, why would they sell the most?
My underlying point was that most popular means only that: Most Popular

"Best" is a quality that is not measured in "units sold".

And now for the transition to UN-biased, factual information:

"That we have the best OS? The best (virus free) user experience? The best interface? The best phone? The best design?"
Ha ha ha, no, no, no, and noooooo. But who is this "we"? Are you in the Jobs family or does giving a company money make you feel like family?


"And Intel made a very wise choice by hooking their wagon to Apple's star."
Now, if we shift focus to reality, here's what really happened. Apple needed Intel so they could run Windows and hence sell more computers (It is one of their primary selling and marketing points). Intel was just happy to have another customer, just as Microsoft is more than happy to sell Windows to new and existing Mac customers.


The only real, consistent success story in the industry for the past few years has been Apple.
Wrong. And now a question: Microsoft has earned an increasing level of profit for how many years in a row now? At least 5... Apple doesn't even come close to that. And given that businesses succeed for the sole purpose of generating profit, your statement is not only laugh out loud funny, but incorrect and terribly biased.

Research first, post second

I agree with most here. Having the biggest amount of money don't grant you all sorts of privileges just like that. You need to invest it, monitor it and take advantage of it.

Microsoft is very good, and sometimes the best at many things. I like the fact that i don't need to wait to download updates until the automatic deployment kicks in.

When i found that they roll out updates, i open Windows Update and check for them. It downloads them right away and i never had any single issue. Single computer or many at the same time.

Is it not surprising that this MS bashing is on the rise everyday. What surprises me is that most articles/notes are so biased and yet people seem to believe/trust them.

Everybody is talking about the fact that Microsoft has the money so it is normal, but don't see that they have the money, and they spend it for this good cause.

100% uptime, yet if I'm doing anything else on my laptop while trying to download updates, I lose connection to the server and the updates fail, so I have to download them all over again when I've finished working

(acxz said @ #21)
100% uptime, yet if I'm doing anything else on my laptop while trying to download updates, I lose connection to the server and the updates fail, so I have to download them all over again when I've finished working :rolleyes:

Thats a problem with your comp not microsoft. I can do multiple things at once while the update is downloading and installing without a problem. So don't go blaming microsoft for your problems.

(majortom1981 said @ #21.1)

Thats a problem with your comp not microsoft. I can do multiple things at once while the update is downloading and installing without a problem. So don't go blaming microsoft for your problems.


A problem with my computer which is running... *drum roll* Windows! So it is their fault Apple's Software Update and Ubuntu's updater can work while I'm doing other things, so why can't Microsoft's?

(acxz said @ #21.2)

A problem with my computer which is running... *drum roll* Windows! So it is their fault Apple's Software Update and Ubuntu's updater can work while I'm doing other things, so why can't Microsoft's? :)

You cannot be 100% sure it is Windows that is causing this. There are MANY other factors that could be the problem. I never had had this issue. More likely whatever you are doing while the updates are downloading is causing interference with the updates. Then your connection to the update server gets lost and you have to start all over again. Blaming MS for this issue is like blaming BMW for your tired going bad cuz you hit a pot hole in the road. Not everything is MS fault

(acxz said @ #21.2)

A problem with my computer which is running... *drum roll* Windows! So it is their fault Apple's Software Update and Ubuntu's updater can work while I'm doing other things, so why can't Microsoft's? :)

You're on a laptop? Not using WiFi, are you?

One other thing that has already been noted, but is still nonetheless valid. . is that Microsoft charges on average $400 for a license for any particular software product, with the price range extending from $100 to several thousand dollars. This being said, smaller companies like Apple or FOSS vendors Canonical have less income to spend on distributed, load balancing and failover servers/networks. Sure, they have smaller user bases, but Microsoft as a larger entity still has an advantage in many respects.

(Divide Overflow said @ #20)
One other thing that has already been noted, but is still nonetheless valid. . is that Microsoft charges on average $400 for a license for any particular software product, with the price range extending from $100 to several thousand dollars. This being said, smaller companies like Apple or FOSS vendors Canonical have less income to spend on distributed, load balancing and failover servers/networks. Sure, they have smaller user bases, but Microsoft as a larger entity still has an advantage in many respects.

Yea, but Apple also charges a lot of $$$ for their hardware and you an only get a Mac from Apple. So they are still making a butt load of money. Also, they have the iPod and iPhone which are doing really well. Granted, they still arnt MS.

Doesn't Microsoft still use Akamai to "mirror" their content closer to the user? If so, then this comparison is very skewed against Ubuntu since their mirrors are not included. Also, if they were only checking every 5 minutes, then their methodology is called into question as well, as there could be many short downtimes that went unnoticed in their survey (for all operating systems).

No, the stats are not stupid. This isnt just for home users. MS having 100% up time is damn good for companies who apply updates automatically on production servers after hours and at odd times during the day. And you have to look at the time zones as well. When it is 12 noon here, it may be midnight some place else and where it is midnight, is when companies are applying updates on their servers/workstations.

I have had problems apply update with linux. Havnt really had any issues with Apple updates tho.

The point is you should be able to apply updates whenever you want no matter the time of day. I work on PCs and servers all the time and when I am building a new machine, I need to know I can access Windows updates at any time.

May have had 100% uptime, but there have been several times I have gotten timeout errors or service not available or some freaked out thing from the site, just in the last month!


When I used Ububntu, i found that the servers other than the main one, were always slow and out dated also.

Yes, VERY stupid. The only time updates are on for me is when I do s sudo apt get or update/add apps. I don't need it to be on 100% of the time

These stats are so stupid. This doesn't affect anyone here. Stupid fanbois foam at the mouth to argue about anything.

(night_stalker_z said @ #14)
Microsoft may have the lowest uptime but the download speed is very slow.
I've hit 5MB/s download on their servers. That's 5 Megabytes a second! They have got some of the fattest pipes my internet can take advantage of

(tsupersonic said @ #14.2)
I've hit 5MB/s download on their servers. That's 5 Megabytes a second! They have got some of the fattest pipes my internet can take advantage of

I got 4.3 MB/s but never past. On average I get 3-3.5 though. On online downloads I get like 800-900 Kbps though, so I'm pretty sure your statement is right.

Good job Microsoft :)

I remembered when Ubuntu 8.04 was released. I couldn't get any updates for it (even using the mirrors). Tsk tsk

christ Neowin .. your becoming Ars puppies

"Microsoft doesn't have much to be proud of, as it's the biggest of the three, and it's no surprise it took first place. Apple's 99.9 percent is also quite respectable. "

100% uptime is hell to achieve. Just because MSFT is huge, doesn't mean they should automaticly be able to turn the world upside down.

I agree. Whoever wrote that obviously has never worked in business continuity or systems administration. Companies strive to have that number. It has nothing to do with how big they are, just that they have some great network planners and administrators.

I would agree as well. And @ that guy that said the chart is misleading, THAT IS THE POINT OF GRAPHS. you can scale it however you want to make it look the way you want it to. Obviously, the point of this chart is to show the difference, not the difference between 98,99 and 100%

(Marshalus said @ #11.2)
Agreed. If I could achieve even 99% during that time my life would be complete. 100% is a fantastic statistic, regardless of who you are.


You can achieve 100% uptime using balancing load and several server if not thousand, Microsoft have the money for do it.

(darkpuma said @ #11.3)
I would agree as well. And @ that guy that said the chart is misleading, THAT IS THE POINT OF GRAPHS. you can scale it however you want to make it look the way you want it to. Obviously, the point of this chart is to show the difference, not the difference between 98,99 and 100% :rolleyes:
A slight correction or clarification to your comment (and the parent who said the chart was misleading).

The chart is accurate, but it is not graphing "uptime", it is graphing "downtime". The scaling they use is appropriate for the data shown.

(Magallanes said @ #11.4)


You can achieve 100% uptime using balancing load and several server if not thousand, Microsoft have the money for do it.

You know its really not as simple as your saying it..

I used to run a private server for an online game and had 2 servers running, and it was hard to monitor just those! Now try 5,000+, I'd love to see you skip through it as you did in your speech.

(Swordnyx said @ #11.6)
You know its really not as simple as your saying it..

I used to run a private server for an online game and had 2 servers running, and it was hard to monitor just those! Now try 5,000+, I'd love to see you skip through it as you did in your speech. :happy:

Ummm... You do realize, right, that "load balancing" servers are not the same as two separate servers?

First of all, canonical isn't as big as microsoft or even apple, which is something you need to factor in here.

Secondly, I would like to know how much traffic these servers were hit with!

Correct me if I'm wrong, I've been out of the windows loop for nearly 4 years now, but I thought the windows update thing is mostly used for small patches & fixes, for bigger microsoft software like IE and for the very big service packs. Right?
And if nobody has been fooling me, microsoft chooses who gets what update at a specific time.

You can't compare that to the ubuntu servers.

The ubuntu repositories are used for EVERYTHING. You can get nearly every piece of software you need on these servers. Of course, there are a lot more windows users in the world, this is still something to keep in mind.

I also believe the biggest hog on these servers is the six-month release schedule. We get a new OS every six months! Windows users don't even get a service pack that often. That's a lot of MB that needs to be downloaded by every Ubuntu user! I agree that the servers go incredibly slow at that time (but I mostly update to the newer release a few days before or after official release)

Ps: Freeware junk, seriously? That's the statement you're making?

Microsoft releases nearly everything through Windows Update, from hotfixes to service packs. They also don't choose what time you can download updates but do phase releases with auto-update to prevent stress on their servers and on the internet in general (hundreds of millions of people downloading Vista SP1 simultaneously would have been unbearable).

So yes Microsoft deserves credit. And yes, Ubuntu needs to improve. However, Ubuntu is free, is a smaller operation and distributes a lot more content. It's Apple that needs to improve most - they put out the signs of being a big player but it's still a lot of hot air.

Wow, that's a misleading chart. On a realistic scale there probably wouldn't even be a difference between the three.

Yeah, at first glance it looks like Microsoft beat Apple and slaughtered Ubuntu.

But then you look at uptime percentage. If you plot 0%, 0.01% and 1.36% on a chart you wouldn't even be able to see a difference.

Still, a perfect score on an update service that provides the vast majority of computer updates is very impressive.

"Microsoft doesn't have much to be proud of, as it's the biggest of the three".

Eh? Serving up terabytes of data to tens of millions of systems with no downtime? Sure, they were cheating, using all that expensive proprietary software whereas Ubuntu had to make do with its freeware junk.

This confuses me, they post about Windows Update beats Apple Software Update and Ubuntu Archive, but then they add this:

Microsoft doesn't have much to be proud of, as it's the biggest of the three, and it's no surprise it took first place. Apple's 99.9 percent is also quite respectable.

Whats the point of this statistic anyway???

(thenonhacker said @ #6.2)
In short: ArsTechnica is biased against Microsoft.

Makes sense on account of its constant flamming.. It is ArsTechnica.

Sorry guys, my statement was supposed to be "Microsoft is the biggest of the three so it's no surprise it took first place."

(thenonhacker said @ #6.2)
In short: ArsTechnica is biased against Microsoft.

LOL. You obviously don't know who wrote this news article.

Microsoft doesn't have much to be proud of, as it's the biggest of the three, and it's no surprise it took first place.

I would say that Microsoft does have something to be proud of. Because it is the biggest also meads it would be the service hit the hardest but yet Microsoft is the only one with 0 downtime.

(bryonhowley said @ #4)

I would say that Microsoft does have something to be proud of. Because it is the biggest also meads it would be the service hit the hardest but yet Microsoft is the only one with 0 downtime.

SK[ said,#2]And then they go congratulate Apple for having 99.99% uptime. Yet another bash MS article.

Agreed. Because Microsoft is the biggest, generally the demand is the highest, meaning the stress on the servers is greater. Kudos to Microsoft.

SK[ said,#4.1]And then they go congratulate Apple for having 99.99% uptime. Yet another bash MS article.

So true. If Apple and Microsoft's data was flipped, they would certainly bash MS there too, saying some BS with it.. I hate fanboys.. I mean, get a life and speak the truth, who gives a care about what you choose?

yeah this is because microsoft couldn't get away with windows update being down. loads of people would start freaking out pretty quickly. with the others it matters a lot less.

It should be noted, though, that Ubuntu’s repositories have mirrors around the world, so users can download packages from those as well....

...For Ubuntu we monitored the URL “archive.ubuntu.com” which is the main repository that contains all the packages that can be downloaded through Ubuntu. Again, it should be pointed out that there are other mirrors that can be used.

http://royal.pingdom.com/?p=319

So, in actual fact, you can get updates for Ubuntu all the time... the data here is skewed.

(Mr Fish said @ #2)
So, in actual fact, you can get updates for Ubuntu all the time... the data here is skewed.

Good point, although they should use load-balancing instead of mirrored servers.

Also, the amount of users for each OS should be taken into consideration, I'm betting MS have at least 10 times more servers for updates than Apple.

(Exosphere said @ #2.1)

Good point, although they should use load-balancing instead of mirrored servers.

Good point, but with mirrored servers I can pick a specific mirror (namely one that is part of the PIPE network here in Australia) and have that download excluded from my monthly download quota Yeah, we have download quotas here on most residential plans..

(Mr Fish said @ #2)

http://royal.pingdom.com/?p=319

So, in actual fact, you can get updates for Ubuntu all the time... the data here is skewed.

Bad point, have you checked the status of their mirrors?! most of them are very slow and always outdated by a big diffrence than the main servers.

(Beastage said @ #2.3)
Bad point, have you checked the status of their mirrors?! most of them are very slow and always outdated by a big diffrence than the main servers.

I have actually never heard of that comment. Most of the mirrors are very fast even when you're using a mirror outside your own country. Most of the mirrors are just as updated as the "main" repository or nearly as updated as the "main" but some of them are not and differ a lot (those are the unofficial mirrors). The official mirrors are hosted by organisations like universities and ISP's. Those organisations have a very good infrastructure and reliability, organisations who are able to realise an uptime of 99,9%. I myself haven't noticed any speed differences between the mirrors in .nl, .uk, .de and the "main" one. Neither have I noticed any difference in the amounts of updates pushed or any big difference between those repositories.

There have been 3 occasions (as in the 3 faulty updates that got pushed out) already where a slight delay in pushing the updates to all the repositories actually saved a lot of trouble. It was big enough to stop the faulty updates from pushing to the other repositories and eventually to the computers. Having a bit of a delay between a complete sync has more advantages than just that btw. There will always be some sort of delay because we have different timezones. People who are trying to update machines will be affected by this and even the mirrors. Timezones are a good thing for infrastructures: it's like a natural loadbalancer :)

I actually have found more problems with Windows Updates not working than Apple or Ubuntu linux updates not working. Windows Updates tends to time out quite often compared to the other two. Apple update can be slow sometimes, especially when a new "service pack" is out. Haven't noticed that with Ubuntu, only when a new release is out. Windows Updates on the other hand is mostly slow and a lot of people complain when it's the first tuesday of the month again. Akamai also notices a big hit on their capacity. This is something I actually can't find in the article nor am I able to find anything about what "a test every 5 minutes" means. There is no explanation what they tested in that 5-minute test which makes the entire test and conclusion worthless. In the end I think it doesn't matter, all 3 systems are equally good but I do think Ubuntu is slightly better because of their mirror systems which is better suited to balance the load (systems like Windows & Apple Update are not like that).

Ah, I think I know why there's a difference: the infrastructure in the USA is really crappy, the EU has a far better infrastructure especially countries like the Netherlands and Sweden. Downloading from some sites in the USA won't exceed 20 kB/s while we in the Netherlands are able to download with 1+ MB/s (some people are on fiber and have something like 20, 30, 50 or even 100 Mbit/s). So I really do think the infrastructure in the USA really has to do a lot with the outcome of the test and the different speeds/uptimes I'm experiencing (since I'm in the Netherlands).

(dyn said @ #2.4)
The infrastructure in the USA is really crappy, the EU has a far better infrastructure especially countries like the Netherlands and Sweden. Downloading from some sites in the USA won't exceed 20 kB/s while we in the Netherlands are able to download with 1+ MB/s (some people are on fiber and have something like 20, 30, 50 or even 100 Mbit/s). So I really do think the infrastructure in the USA really has to do a lot with the outcome of the test and the different speeds/uptimes I'm experiencing (since I'm in the Netherlands).

There is this big thing called the Atlantic Ocean, it tends to slow down Internet traffic a bit.

I have little problem getting fast speeds in the US, but I'm from the US. I've never really had too much problem getting decent speeds from the US to the EU either.

Agreed that this is horribly skewed against Ubuntu by artificially limiting downloads only from the "archive.ubuntu.com" main site, and not any of the local mirrors as well.

Congrats to Microsoft on their 100%. Apple did pretty well, too. And shame on Pingdom for even thinking that this could be promoted as a valid comparison.