Articles:
I've previously spoke about Converged Infrastructure and its readiness for hosting business critical systems which were previously the domain of Mainframe and Open Systems (UNIX), I've also spoke about Cloud Reference Architectures and how these can ease the adoption of Cloud Computing within your organisation. In this article I want to talk a little about deciding which workloads or applications can easily be migrated to Cloud Architectures, based on Converged Infrastructures, and why this will benefit your organisation.
Many IT departments are struggling to deliver increased benefit to their businesses whilst reducing their budgets and maintaining or improving service. I see this everyday when talking to senior IT leaders in both the UK and wider EMEA countries. Everybody is struggling and given the current economic climate, this really isn't surprising.
One of the strategies I advocate and I'm happy to say one which is being adopted by organisations both small and large from all sectors is the migration of expensive proprietary UNIX systems to cost-effective x86 (Intel) based systems running Linux. Many IT leaders already know that this makes sound economic sense and there are many simple cost models which can be used to prove this.
The stumbling block is more often than not understanding the approach that needs to be taken to identification of migration candidate workloads so being able to understand areas such as application dependencies, current infrastructure utilisation, transformation and consolidation opportunities. Another much more practical question that I am asked is around people and the training and skills required.
So how are we helping to educate IT leaders and provide them with the information they need to be able to make a well informed and decision to help deliver the increased benefits and competitive advantage so many businesses are no looking for their internal IT to deliver?
Your first step should be a rigorous discovery exercise that will enable you to understand all of the physical and virtual infrastructure you already have deployed in your organisation. On top of this you will need to layer your service technology patterns, which should include any application interdependencies before finally understanding the workload performance and utilisation metrics.
Once you have this information you can begin the process of classification. This can be as simple or as complicated a process as your organisation demands.
I have come across both ends of this spectrum from simple;
- is it Solaris & Apache on SPARC? then it goes to Linux & Apache on x86 (Intel)
or
- is it HP-UX and Weblogic on PA-RISC? then it goes to Linux & SpringSource on x86 (Intel)
to a more complex decision tree;
- is it HP-UX & Oracle on PA-RISC? then understand I/O profile, Storage Requirements, Business Criticality, Be-Spoke Scripts and Disaster Recovery
- is it AIX & SAP on POWER? then go through a complete business case process
However, it is fair to say that I have only come across a hand full of instances whereby the decision to remain on proprietary UNIX has been overwhelming, and most of those were UK Central Government critical systems.
The final key aspect around people and their skills is surprisingly variable between individual organisations. Many people know and love the x86 architecture and Intel in particular as they have these on their desks at home in the office and more and more on the move. In particular UNIX administrators will more than likely have multiple x86 based machines running a plethora of Operating Systems, including Linux, often in their lofts or garages. You see UNIX administrators are techies and I think I can safely say, the worst sort (I say this having being one myself). They are often very technically aware and will often have very complex home networks and will consume IT whitepapers, Redbooks and code throughout the day.
They know how to use Linux, they might have a preference in terms of the distribution they use, but the underlying LSB that most popular Linux distributions are now built on and the heritage of the original GNU code base and Linux Kernel means that they can leverage their UNIX skills; be they AIX, Solaris or HP-UX.
This is where the 'worst sort' comes in.
Some UNIX administrators are very resistant to change. They have spent many years developing ksh, bash or C scripts to help run their UNIX estate. They have often developed functionality prior to it being adopted in main stream proprietary UNIX distributions and incorporated it into their UNIX estate. They see your business services as their UNIX estate and if this is the case you will find it very difficult to gain migrate to the more cost-effective x86 based Linux solution that I advocate in my daily interactions with IT Leaders.
‘For more information please click - HERE’
‘This post is brought to you in partnership with Intel(R) as part of the "Technology in tomorrow's cloud & virtual desktop" series’
More Articles...
By A Web Design
Loading...
Last 2 tweets from openalternative:
Conducting your Cloud
A fellow blogger suggested blogging on what had happened to me over the previous week, and although getting drunk and eating too much would probably make a great post, it doesn't quite fit with this site,...
Cloudy with a chance of Blue Skies
The current global economic climate is a major cause for concern for all industries and individuals alike. We haven't seen a global recession the likes of which we are currently experiencing since the...
Migrating from UNIX to Linux in the Cloud
I've previously spoke about Converged Infrastructure and its readiness...
Cloud Reference Architecture
So what is a Cloud Reference Architecture?
When you consume services from a Public Cloud Provider do you really care or want to understand what the underlying architecture looks like? Probably...
Who's Online
We have 2 guests and 2 members online