We are sure happy to welcome GitHub to Rackspace. I have gotten to know the founders of GitHub over the last year, and there is no question this is a team and concept that the world should not ignore. We are excited to support them as they continue to deliver great services to the developer community.

It goes without saying that in reaching out to them one of my intentions was to get them on our cloud.

I was surprised to hear a few months later that they were eager to move to us, but they didn’t want cloud, they wanted primarily dedicated servers. Here is true leader of the new era of computing, a team born in the cloud era, and they wanted dedicated gear? What gives?

Well, here is how they describe it in their blog:

“We’ve grown to a size where it no longer makes sense to have every server virtualized. The benefits of running bare metal are obvious and have been empirically proven. We need to have the option to run bare metal when it is appropriate to the task at hand. We also need to be able to configure boxes with custom setups.”

There are really two cloud debates going on right now. One, is simply about the era of buying computing over the web vs. building it in house. This is what we call the cloud movement and it is real and has serious implications. We truly believe 99% of companies in the world can move faster, save money, and perform their IT better by using a computing partner. The next debate is about the specific tool now being called “Cloud.” These are pooled services, powered by software allowing for real time provisioning and very granular pay for use. We think of these as cloud technologies. All this jargon and cross talk has caused much confusion.

We think the Github decision is a great example to use to try to share our views on it. So:

One, we do think most companies going forward will buy computing not run it themselves. No capex, no inventory, no lock in. Github is not interested in having a DC and all the headaches that go with it. In fact, they have never really even considered that option (what startup today would?). We are discussing this movement daily on our site nomoreservers.com.

Two, while this strategy could be called using “the cloud” there is no doubt in our mind that companies will choose from a variety of services based on the workload they are dealing with. They will not just use Cloud technologies like our Cloud Servers or Amazon’s EC2. They will use traditional technologies as well as even higher level services, like SaaS. Why?

Well, here is how we think about it:

If you see these tools as a continuum of prescribed technology deliver on demand, they set up a series of trades. The trades are pretty simple:

The Github team faced this exact trade. As they grew, they needed the ability to customize every aspect of the stack by using physical machines. This is not to say they don’t believe in the Cloud technologies. Later in their post they say:

“On-demand access to a cloud infrastructure will be important to us as we increase the number and variety of low-frequency but long-running jobs that we process.”

The “Cloud” is a part of the computing strategy for Github. As it should be for any company. There are workloads where it is a perfect fit. Other workloads will need other tools. So, as the cloud era picks up steam, the discussion about all or nothing decisions of using one tool or another should be ignored. The key for any IT department is to figure out where the tools match their needs based on the workload they are deploying. Follow the lead of Github.

Rackspace held a System Administration breakfix contest at the 2009 Red Hat Summit in Chicago pitting the top Red Hat Certified professionals against each other for the chance to win prizes and win bragging rights. During the two day contest, forty-seven participants worked to solve the four identified “breaks” within the fifteen minutes allotted. The problems, listed below, were chosen as representative of issues Sys Admins in a hosting environment must be able to resolve on a daily basis. The contestants who resolved the most problems in the shortest times respectively won first, second and third place.

With encouragement from @devrandy, Randy Russell, the Director of Certification for Red Hat, there was quite a heated competition among the contestants for the top three spots. The winners respectively won one of three prizes: 1st Place, an Acer Aspire 5155; 2nd Place, $100 ThinkGeek gift certificate; 3rd Place, 8GB USB drive. Our winners were:

1st Place – Alex Davies (also a Red Hat 2009 RHCE of the Year!) – 4 correct solutions in 12 minutes

2nd Place – Hai Wu – 3 correct solutions in 15 minutes

3rd Place – Dave Johnson – 3 correct solutions in 15 minutes

Of the forty-seven contestants, only Alex successfully solved all four problems in the allotted time. Thirty-two correctly solved Problem One. Eleven solved Problem Two. Three solved Problem Three and four contestants solved Problem Four. Congratulations to our winners!

Rackspace Trainer Mike Roberts (@soopurman on Twitter) provides a multi-part screen cast of the four challenge questions and provides insight in troubleshooting and solving each. Below, Mike’s screen cast is broken into segments for each problem in the challenge. These are interspersed with the actual questions from the breakfix challenge for reference. Mike begins with a brief introduction to the breakfix:

Rackspace Breakfix Challenge Instructions

This exercise is meant to approximate the types of problems that Rackers solve for our customers with Fanatical Support every day.  As such, the idea is that you should attempt to detect and correct any possible misconfigurations, but not drastically alter the basic setup, nor replace the actual services being used.

We will give you up to 15 minutes to complete the following problems. When you are finished, ask a Racker to record your name, the problems you fixed, and your completion time.  On Friday we will award prizes for the fastest successful completion of all challenge problems.  Good Luck!

1. The IP address 192.168.0.253 is accessible by others on the network, but this machine cannot succesfully ‘ping’ it.  Figure out why and fix it.

2. This machine is supposed to offer both anonymous vsftpd and anonymous rsync services, but clients are complaining they cannot connect to either one.  For example, the command “rsync rsync://localhost” should display a share called “challenge” but it does not.  Figure out why and fix it.

3. The default page of the web site running on this machine is supposed to simply display the word “success” but it does not. Figure out why and fix it.

4. The password for the user named “rack” is correctly set to “rackspace” but this user still cannot login.  Figure out why and fix it.

Mike concludes the presentation of the solution to the breakfix challenge with a special invitation to System Administrators who may want to pursue this kind of activity on a regular basis:

For more information about Rackspace and available careers  please visit our Career portal: http://RackspaceCareers.com.

Looks like quite a few people are digging what we’re doing here @Rackcloud.

Recently, the web research shop InfiniBase wanted to know who the leader of the pack was in cloud providers.  The numbers hadn’t been run yet, so they examined more than 500,000 sites to discern what companies are in the cloud and which companies host them.   We are stoked that Rackspace is a “dominant” force in the cloud market.  The full article can be found here.

Do you want to deploy some form of ‘Rich Media’ online? With so many things to consider, first and foremost on your mind is probably HOW: host it yourself or outsource it? How much will this cost? Do you have the right people to make it successful?

Asking and answering the right questions sets you up to succeed; because a failed or broken media campaign can be far more damaging than no campaign at all.

Three key factors to a successful online campaign of any kind are delivery, availability and performance. Learn more about how to launch your rich media campaign here.

In this ongoing series, we will help you ask and answer these and many other questions.

Erik Carlin wrote a great post on the Mosso blog comparing the cloud storage offerings from Rackspace (Cloud Files) and Amazon (S3). Here’s a quick excerpt:

The cloud is advantageous for many reasons and both Rackspace/Mosso and Amazon offer cloud storage solutions.  We are frequently asked to compare Cloud Files enabled with Limelight’s CDN with S3 and CloudFront.  Many of the questions we are asked revolve around cost and performance (particularly CDN).  These are very quantifiable metrics so I thought I’d share with you the results of some comparative analysis we’ve done.

Read the entire post here: A Quantitative Comparison of Rackspace and Amazon Cloud Storage Solutions

These days, it seems to be all about the cloud.  In the last couple of years, this new variant of hosting has built into quite a storm (pardon the pun) amongst the IT community.  On whiteboards all across the world, amorphous blobs that used to represent what lay outside the application infrastructure ( the Internet, for example) have now been elevated to part of the design.   While the change may seem subtle in regards to brainstorming and spit-balling your way through a design session, the options that cloud hosting brings to developers is quite significant.  The challenge, however, is making sense of all the options and how to leverage the different variants this new trend provides.

Cloud can mean a lot of things.  Sometimes it refers to a storage option.  It can also describe a self-contained application platform.  The term is even used to reference leveraged virtual machines where the customer doesn’t have to worry about any of the underlying physical machines.  With all the rapid innovation in this area of the hosting industry, it’s easy to see where people might start getting confused.  While more and more IT shops have embraced various cloud hosting options as part of their solutions, it’s clear that there is a huge population of enterprises that haven’t taken advantage of these scalable and flexible offerings – primarily because they don’t have a comfortable level of understanding.

This became remarkably clear when Rackspace® Hosting recently commissioned a study by by Vanson Bourne to gauge the level of awareness about cloud hosting among businesses of varying sizes in both the US and the UK.  Earlier this week, we released the findings of that survey.  The statement said that while: (read more…)

Every time I talk to customer that is joining the Racker family or run’s into an issue with scaling their IT infrastrcuture, I find that they either don’t know how backup and storage needs will effect their business or how to best enable today’s storage product’s to better enable their business needs. It seems that when it comes to architecting the largest most bullet proof solution in the world either the storage portion or the backup solution are thought of last. Don’t get me wrong, I can understand why this is the case and in most case understand why it is the last thing to consider. However I would like to recommend a new way to think about architecting your solution when storage or backup is needed.

The new approach I like to take is what I call “Block-up” infrastructure planning. Block-up planning is a very easy process. You first decide how much data space you need (in term of GB), what do you need it for (In terms of a business application, like a database) and how much this data is worth to your company/business? I will give you an example of this and how this is important. (read more…)

Greetings from San Francisco! I’m here because there are two conferences currently occurring that we think you’ll find interesting.

The first conference is one you may have heard of – the 10th annual LinuxWorld Conference and Expo. Each year, around 10,000 software developers, system administrators and IT professionals attend LinuxWorld, making it the largest conference focusing exclusively on Linux and other open-source technologies.

The second conference actually spun out of LinuxWorld two years ago – the Next Generation Data Center conference. The two conferences share a lot in common – both shows are produced by the same company, they take place in the same location, and they actually share an exhibit hall. If you didn’t know better, you’d think that it is just one big conference – that is, until event security tackled you for trying to get in to one conference when you only paid for the other.

Luckily for you, I’m going to give you the run down on both conferences – thus eliminating “dodging event security” from the list of things you need to worry about today. You can thank me later.

If you were at either conference, chances are that you ran in to someone from Rackspace at some point. Rackspace CTO John Engates presented at both conferences, and Mosso co-founder Jonathan Bryce is presenting at LinuxWorld on Thursday. We also have an awesome booth where we are running a server break-fix competition, as well as a booth in the career fair pavilion where we are looking for the next generation of Rackers. Everywhere that you turn, Rackspace is there. We’re subtle like that. (read more…)

DNS or Domain Name System is a service most commonly used to translate Domain Names (URL’s, websites, call them what you will) into IP Addresses. Realistically, IP Addresses are the true identifiers of how to locate things/places on the Internet, however DNS makes it much simpler to “surf” by only having to reference Domain Names that are much easier to remember. DNS functions in a hierarchical structure

Rackspace uses an implementation strategy called AnyCast with our Authoritative DNS. This technology allows us to announce the same DNS IP space from 3 different (or as many as desired) datacenters via the BGP protocol we run with our Internet Service Providers. The purpose of this is multi-fold in that it not only allows us to have active/active DR type redundancy, but it also allows us to serve customer requests from the closest DNS infrastructure possible, which typically means the best response time too. Additionally, it allows for the ease of maintenance, upgrades, and expansion our our DNS infrastructure with no customer impact.

This is the same kind of underlying technology that the Root Nameservers of the Internet run off of, and many other companies that specialize in DNS as a service.

Today we announced the private beta of the much anticipated CloudFS, an Internet-based storage offering that was developed by Mosso and Racklabs. I’ve included a couple interesting excerpts from the press release below:

This file storage offering, CloudFS, allows developers to securely store a virtually unlimited amount of data on the web connected through Rackspace’s infrastructure. At the conclusion of the beta program, the new storage offering will be available as a stand-alone service or as part of Mosso’s cloud hosting solution at competitive storage, bandwidth and transaction pricing. In addition, Rackspace customers will benefit from free local transfers between CloudFS and their managed hosting servers. The ability to tie a scalable and low-cost storage engine to Rackspace’s customized managed hosting environment is unique in the hosting industry.

Mosso is seeking applicants to participate in a private beta program as well as encouraging developers to build tools and services around CloudFS. Developers and businesses interested in the private beta can visit www.mosso.com/cloudfs to apply starting May 5, 2008.

Pricing for CloudFS is strictly pay-for-what-you-use. Current plans target the pricing for storage to start at $0.15 per gigabyte, upon release, including replicated copies for data protection. This competitive rate allows developers to scale their storage-related projects seamlessly and affordably.

The storage platform will be accessed via a ReSTful web services API and language-specific API’s (.NET, Java, PHP, Ruby, Python). CloudFS will support language-specific libraries, giving customers confidence that they can quickly and easily store data and run applications in a business-class, standards-based environment.

Check out John Engates’ post on the Racklabs blog for an interesting behind the scenes look at the development of CloudFS.