From Servers to Services. The Role for the IT Pro in the Cloud.
The looming question for IT Pro’s regarding the cloud is “What is the future for my role?” Simply put, no one can say for sure. There are many factors to consider when answering that question. The cloud offerings are not identical, which affects the answer. There will be changes in roles. That’s not unusual for us IT Pro’s, and these changes probably will not be universal across all organizations.
So what are the potential changes for the IT Pro? Let’s look at what cloud offerings change in our organization, starting with Public Cloud. In a nutshell, the most obvious change when employing any public cloud solution is the sudden disappearance of servers. Whether it’s a BPOS solution, Windows Azure, Amazon Web Service, Google or Salesforce, the physical “tin” in our world goes away. If there are no servers does that release us from management duties? Well, yes and no.
Let’s drill one level deeper and consider SaaS, PaaS and IaaS. BPOS, as we’ve previously discussed, is a SaaS offering, so if you deploy Exchange Online in your organization the Exchange servers potentially go away. We’ll keep the scenarios simple and go with a complete cloud implementation of Exchange Online, but the same applies to SharePoint online and Office Live Meeting. The underlying management of servers, patching etc, is now handled by Microsoft. You can still maintain the directory service whether on-premise via AD and synchronised to Exchange online or via the Exchange Online management interface. As a footnote, along with the directory you can also synchronise on-premise Exchange servers with Exchange online. The Service Administrator for Microsoft Online Services has a number of tasks that IT Pros must perform. Many of these tasks are similar to administrative tasks IT Pro’s perform for a fully on-premise solution. The part of the role that increases in importance is managing the service provider. We all deal with ISP’s for our Internet connectivity; occasionally we lose the service and escalate issues to them. Often, this service loss would not affect our messaging infrastructure to the degree that mails cannot be sent. With Exchange Online that is not strictly the case. Monitoring and escalation of issues become more critical. Understanding of the SLA’s is important, understands how to pinpoint the issue, and know which suppliers’ SLA an outage falls under.
What about PaaS? Solutions from Google, Salesforce and Windows Azure also remove servers from your organisation. Like SaaS above, there is no server patching. In fact, SaaS and PaaS solutions are similar. The main difference is that the maintenance of the application remains your responsibility. Windows Azure Platform takes care of everything in the stack up to your application. From that point on, application failures, application updates or upgrades are still part of the role. Also, billing comes into play. Because PaaS solutions are pay-as-you-consume models, judging when to increase capacity and when to decrease it becomes important. The idea of cloud computing is to provide massive computing resource to organizations but to pay only for what you use, so leaving a few instances of your application running while unutilised can be wasting of money. So, keeping an eye on utilisation and being able to judge capacity becomes an increasingly useful skill. Other skills that are important are backing up your data. Remember your data is now sitting in Azure storage or SQL Azure; if you want copies you have to copy it to your local storage. You will need to acquire skills in managing security using the Windows Azure platform Appfabric. Also, when using AD FS 2.0 to manage access you should be aware of “claims” and how this access control system works.
Finally, IaaS! With IaaS you loose the hardware but retain the server management. For example, with Amazon Web Service, you run a virtual machine on their hardware infrastructure but you are responsible for the OS running on that virtual machine. So, both patching the OS and any application running on that VM are your responsibility. IaaS requires all the skill you have today, plus some additional responsibilities for managing the provider. We could have the debate about whether IaaS is a very similar model to outsourcing your infrastructure to a hoster. With the Dynamic Datacentre Toolkit for Hosters, Microsoft enables hosters to build private and public cloud offerings. IaaS becomes a very interesting area of cloud computing as traditional web hosters could enter the cloud market and offer to host more than your web servers. They could provide an additional service in keeping your machine updated with critical patches and service monitoring. With this in mind, it’s not hard to see why some leading experts are predicting that data centre infrastructure could be all in the cloud by 2015. I was talking about this very point recently and the question about the IT Pro role in this scenario. It dawn on me, not only do I have a good answer, I was experiencing a similar scenario today. I mange some production servers, but I’ve never seen them. Never been near them , could not even describe what they look like. I perform all the management tasks using Remote Desktop; you could say that I’m using an IaaS type service. I rely 100% on Microsoft Datacentre support to manage the environment and the hardware (server and storage), but everything beyond that is my responsibility. This begs the question; if my servers were running on Windows Azure or Amazon Web Service instead would I notice the difference?
It would be remiss not to mention private clouds. As an IT Pro these can be considered in their simplest form of racks of servers running, say, Windows Server 2008 R2 with Hyper-V, managed by System Center Virtual Machine manager. With private clouds, as the name suggests, organizations own and run everything on-premise as they do today, with traditional servers. Private clouds are a service delivery paradigm change. The role of an IT Pro changes very little in this paradigm, but our ability to deliver a different level of service to our organization does.
This wave of Cloud computing is the next evolution. I say “this wave”, since cloud has been around a while, but it’s only now that huge amounts of resources are being pumped into it. As IT Pro’s we will adapt as we always have. We will gain new responsibilities and lose some old ones. This is an exciting time for IT Pro’s with great opportunities for to change organizations perceptions of the IT Department from a “necessary expense” to an agile operation delivering new innovative services. Whatever happens in the next few years, the world will always need IT Pro’s.
Comments
Anonymous
June 14, 2010
Hey Alan, your article finishes with "the world will always need I.T Pro's ", but you were to coy to say "but probably less that 50% of who we have now so you will probably lose your job" ! be brave and say it !Anonymous
August 20, 2010
Why do pros put an apostrophe on pros when there shouldn't be one? You create a plural by adding s not 's.Anonymous
August 20, 2010
There shouldn't be an apostrophe in ISPs either.Anonymous
August 20, 2010
The comment has been removedAnonymous
August 20, 2010
There shouldn't be an apostrophe in SLAs.