What every business wants and strives for is IT Service Availability at lower cost. But how you achieve this is a challenge that really concentrates the mind. Moving forward, a triple alliance between Physical, Virtual and Cloud will be the route that most organisations choose to deliver against their availability goals. The big test is deciding which service you should put into what ‘bucket’ of this alliance.
Let’s start off by thinking about business-critical services. These require availability and protection at all times, regardless of whether they reside on a physical server, a virtual machine or potentially both. It is easy to create new virtual machines; it’s not so easy to manage them when too many are generated. On top of that, what are the aggregated server management costs? While the lure of life in the Cloud will supposedly remove capital investment, who actually pays the ‘Cloud bill’ at the end of the month? Getting the balance right in all of this is a complex challenge. However, handled properly, it can deliver the most appropriate or ‘best-in-need’ economics, functionality, security and availability to be had. So it’s a challenge that’s well worth taking on!
One question that often becomes a stumbling block is where each of the aspects of the alliance begins and ends. However, that might be the wrong question, as none of these is discrete, but overlaps/integrates with/supersedes the other along the way. The better question to ask is: “How do I get all elements to ‘talk’ to one another in a way that ensures the triple alliance is putting me ahead of the game and delivering fresh opportunities to give new levels of business agility?” Most importantly, the truly successful organisations achieve that journey – from legacy discrete systems to a dynamic and agile IT environment – through a phased, rather than fast-track, approach.
For example, research from VMware – with whom Symantec has a close partnership – shows that organisations tend to choose one of two paths when they have virtualised about a quarter of their IT environments. The virtualisation of that first 25% will probably include easy choices, such as test and development environments, etc. Once the low-hanging fruit has been harvested, they then must make a choice – and it’s a critical one: “Do I continue down the path to nearly 100% virtualisation?” Or “Do I go beyond virtualisation alone and incorporate cloud-scale operations as the next step in my virtualisation and IT strategy?”
Take the first path and you will be in a position to enjoy the many gains that virtualisation can bring. Take the second and you will not only be able to amplify the benefits of virtualisation, but also take your business agility to new levels, as you further reduce the start-up IT investment required for new business projects.
This ‘25% virtualisation’ point can often become something of an obstacle, as there are other factors that raise concerns and questions about the best way to move forward:
- Am I really in a position to start to move my most critical systems into a virtualised environment (let alone the cloud)?
- What are the other considerations I need to take into account that might impact availability, performance, security breaches, and mobile access?
- What is the actual service level the business needs? This ultimately will dictate the cost of the service
- How do I control and migrate systems from the physical to the virtual and then into the cloud? More importantly, how do I get them back under my domain, should things start to go wrong?
One of the areas into which Symantec is currently putting considerable engineering effort is helping customers move beyond that 25% barrier, building reference architectures (based on TOGAF principles) to define what the requirements are for a mission-critical, virtualised environment. These are just as applicable to someone building their own virtualised infrastructure as they are to someone who wants the assurance that their cloud service provider has built their service based on best practice.
Using a block-based approach to create these reference architectures, it’s possible to see how we can get the elements of the triple alliance to link together. For example: My core Infrastructure is based on a virtualised environment, which plugs into my physical server (or appliance) based back-up environment, which I complement with a cloud-based archiving solution.
By breaking down the basic components or building blocks needed to deliver business-critical services, you can start to understand which elements of the overall service are best met and delivered from which ‘bucket’ of the triple alliance, and so get to that ‘best of need’ service delivery model that the business demands.
Of course, it’s all very well offering you my own insights into all of the benefits that the triple alliance can offer. What I’d also like to know is whether this matches your own experience, in terms of the feedback you are getting from your client base. Are they saying much the same thing? Where do they stand on Physical, Virtual and Cloud? It would be good to hear your thoughts.
Finally, you might want to take a look at this site:
This explains how virtualisation of business-critical applications can be accelerated by increasing visibility and control across application tiers, improving storage I/O performance and availability, and unifying backup and recovery across physical and virtual domains – all while maintaining security and compliance you can trust.