We are still in the midst of cloud hype. I still see far too many vendors taking their existing product range and creating a cloud message around them, much as when VendorA with ProductB became VendorA.com with eProductB during the dotcom boom.
However, just messaging an existing product as cloud does not make it so, much of what is there is just hosted software, not SaaS; virtualization, not IaaS; statically resourced, not elastic. Who are prospective cloud customers to turn to for advice that is a bit more independent?
A little bit of motherhood and apple pie — cloud is a journey, not a destination. Many vendors take the archetypal approach of “if you want to get there, I wouldn’t start from here”, looking to use fork-lift upgrades from everything that you have in place to something new — which is generally, much to the surprise of the customer, the vendor’s products.
No, what is really needed is someone who can look at what you already have, along with figuring out all the dependencies between the various applications and the hardware they reside on. Then, based on understanding the organization’s own risk profile, this chosen independent partner can help define what the desired end result should look like — and the steps that are required to move from what you already have to get there.
Through this mechanism, there is less emphasis on change for change’s sake; more on optimizing existing investments where they make sense, and on proceeding at a pace that makes sense for you, the customer.
The approach that I recommend is one where the initial asset discovery stage looks at the hardware and software, and (as stated above) looks at identifying the dependencies between these. Based on the process flows already in place within the organization, gaps in how well optimized these flows are can be relatively easily identified at this stage.
In many cases, a more independent partner will be able to look at applications and understand that whereas a vendor with an agenda may just be of the mind of “Surely not — still using a mainframe? With COBOL????”, the fact that the application is doing just what the organization needs, is well supported and managed, and looks like it will continue to do so for a good few years means that this is not a “legacy” environment — it is a valid, valuable environment.
There may be other applications which are just about OK — they will definitely need changing in the next 2-3 years, but at the moment, they are not a constraint on the organization’s business. These can be flagged, and a strategic plan put in place for how to update these over a period of time.
Then there are the main priorities: applications that are already a constraint to the business. Surprisingly, many of these turn out to be relatively modern; applications that have been created to deal with bandwagons as they go past — mobile applications that only work on iPads, business intelligence solutions that are still dependent on the IT department creating new reports for users, and so on.
The partner can then work with the business to understand the real feeling about cloud — is there a real perceived concern about data security and the cloud? If so, even though the perception may be wrong, it may be difficult to sell a cloud solution at this time to the business. Where cloud is an option, then the partner can make sure that it is the right platform, taking into account latency, availability, ongoing costs, and so on.
However, any decisions that are made to change any part of the IT platform, require the long-term view to be maintained, and cloud should not be completely ruled out.
Through an approach that is not tied to any particular vendor and looks at the existing and required future states, a reasonable gap analysis can be carried out and suitable advice provided to the business, that can then be taken or left as the business sees fit.
By taking a technology agnostic view of its customers, a good service partner can provide advice that is more open and provides better bang for the buck. From my point of view, this approach is the best way to make sure that cloud works for an organization, and that vendor hype does not cloud reality.
Image credit: stallio (flickr)