Did you ever wonder how busy the servers of the world’s largest social networks are? It turns out it’s very hard work being popular, especially for the number one player.

According to data from Google, Facebook serves 260 billion page views per month. That’s more than six million page views per minute, or a staggering 37.4 trillion page views in a year. We can safely assume that Facebook’s web servers aren’t getting bored waiting around for work to do.

Read more

When Apple introduced push notifications to the iPhone this summer, we immediately thought of what a great match it would be for our monitoring service. After all, the main point of the Pingdom uptime monitoring service is getting alerts when your website or server goes down.

So, when we started work on version 2.0 of our iPhone app, this was one of our top priorities. Now the new version is here, with push notifications, graphs, and more.

Read more

Tomorrow is System Administrator Appreciation Day, which means one thing: Be nice to your local system administrator. Do something nice. Get them something special. Have a party. Get them some insanely cool hardware.

Your system administrator probably deserves more respect than he (or she) is getting, so give them this one day of the year to feel really special and appreciated.

Read more

The newly released Wolfram Alpha is a great tool for doing calculations and data conversions, and it also has a significant amount of data that you can play around with. This post is about how webmasters and sysadmins can benefit from this new service.

So if you’re a webmaster or sysadmin (or just an Internet geek), let Wolfram Alpha’s two supercomputers do some work for you and make your life a bit easier.

Read more

We’re closing in on 500 articles here in our blog (this is number 498), and a fair share of these have been articles that are useful to webmasters (including bloggers). Therefore, here is a selection of our best and most useful articles written with webmasters in mind.

To help you out we have sorted the articles into groups. The index is at the top of the post.

Read more

Is SaaS (software as a service) a trend that is gaining more and more of a foothold in IT departments, or is it doomed to be the bastard stepchild of traditional software? Jeffrey Kaplan from Computerworld set out to debunk five common myths about the SaaS model. He had several interesting points.

Read more

We got tired of not having a good cheat sheet at hand to convert uptime percentages to downtime and vice versa, so we made one. Hopefully you will find it useful (I know we will).

You can download the PDF version here.

Print it out and use it as a reference when you need to quickly check how much downtime in a month 99.5% uptime actually allows for (just an example). It’s very handy. All information on just one page!

You’ll also find a smaller version inside this post, for your convenience.

Read more

When it comes to the ability to do damage to a company, few employees have more power than sysadmins. Deep system access and inside knowledge is a necessary part of their job, but when things go bad between employee and employer, some very sensitive situations can arise.

Here are six real-world cases of “sysadmins gone wild” that all ended up in court.

Read more

We thought it was time for some fun of the geekier kind. If you know what IPv6 is, this should be something for you.

You may have seen IPv6 addresses that contain a couple of actual words. Here is a made-up example: babe:f432:42aa:8271:eee6:1076:dead:beef

Now what if we take this one step further, and construct entire sentences inside IPv6 addresses instead of just a few words? We decided to do just that, and here is how we did it.

Read more

In November, Google wrote in their official blog that they had done an experiment where they had sorted 1 PB (1,000 TB) of data with MapReduce. The information about the sorting itself was impressive, but one thing that stuck in our minds was the following (emphasis added by us):

An interesting question came up while running experiments at such a scale: Where do you put 1PB of sorted data? We were writing it to 48,000 hard drives (we did not use the full capacity of these disks, though), and every time we ran our sort, at least one of our disks managed to break (this is not surprising at all given the duration of the test, the number of disks involved, and the expected lifetime of hard disks).

Each of these sorting runs that Google did lasted six hours. So that would mean that hard drives would be breaking at least 4 times a day for every 48,000 hard drives that a data center is using.

Read more