CTO Blog

Bit of a disjointed series of connections; first I read how a single major trunk connection failure in India takes out a significant amount of the Indian Software Industries online capabilities until it's fixed. Then there has been another Blackberry outage in the USA affecting around 50% of users, plus a few other less headlined stories, all of which can be summed up as extremely large scale system use involving numbers and ‘infrastructural’ services that are simply beyond what we have normally been contemplating in system design. Whilst the Indian example is pure ‘infrastructure’ as we normally define it, the Blackberry problem may, or may not be, but to the users I doubt if they will think of their devices in terms of an application.

We now need a new term for a more complex set of capabilities that extend outside our organisation as well as inside and are the basis for an increasing amount of our personal work, and our enterprises capabilities to do business. My colleague Ron Tolido coined the term ‘Invisible Infostructure’ which to me is a good description. And he defined it as "Using virtualization and grid technologies to deliver infrastructural services – including all facilities to exchange and process (inter) company information – as a commoditized, preferably almost invisible utility". Eventually, even core business services will merge into it, creating a true ‘business infrastructure’. An invisible infostructure captures and supplies information as if it were via the other markets.

In my research and thinking I have been approaching this not from the direction of the failures that I just outlined, but in trying to determine the value proposition of the new wave of technology so that a definition for a design, implementation and operation can be arrived at. I suspect a number of us are haunted by the past when we failed to see the bigger impact of a clutch of technologies as an environment change and provided solutions to elements that quickly became limiting factors. This has lead me to ‘Cloud Computing’ which if I were forced to make a distinction of the term over Grid computing it would be that since Globus started to the drive towards creating high performance computing structures by ‘grids’ of computers we have seen a big change in the technology world. The starting point was from a few users/applications wanting to use many computers in parallel to solve something big, i.e. Research centers etc. sharing resources to utilize them better; to many users doing many small tasks simultaneously with no discernable constants for the allocation of resources.

Well that’s my personal justification for the name change – it’s a debatable issue and your thoughts and comments are welcomed. Look up Wikipedia on this and they add a further interesting definition; ‘Cloud computing is a general label for the many different approaches to the use of shared computing resources, rather than having local servers, or personal devices handling users’ applications’.

The more research I do on Cloud Computing, and the more I hold what I find up against my efforts to define how people are already using technology personally, but how I expect Business use to expand this the more I realize just what a huge impact this will have on our existing ‘infrastructure’, and if we don’t move to change how we will find ourselves the weakest, slowest or least reliable link in any business transaction. Put another way if our email server is unreliable then right now we know it has a big impact internally and frequently externally, but place yourself into Web 2.0 style shared collaboration, or host a community, and it’s going to get distinctly ‘personal’ if you can’t handle the demand.

Reverse that and be the company that can provide ‘smart services’ to extend capabilities to do business better with your customers and suppliers and you have a competitive lever. The challenge is that you have no idea what the demand will be as the skill of determining and sizing ‘online’ interactions is very hazy, and when the interactions relies on services from several resources being combined it gets to be impossible, hence the need for an ‘infostructure’ approach.

Actually the European Union believes in this so much as an enabling factor for business across the EU that it's investing in a 17 million Euro project called ‘RESERVOIR – Resources and Services Virtualization without barriers – with IBM acting as a lead for what they refer to as ‘the internet operating system for business’. Meantime at a more every day level in addition to IBM there is HP, Sun and even Amazon all out there offering what, to judge from the take up, are pretty popular early variations on Cloud Computing, though it might be better to describe them as ‘utility computing’ though Amazon of course call it ‘the elastic cloud’. And now I see that Dell is joining in as well.

Add a Green IT dimension to this as well and it really does seem that the time has come to do some serious re evaluation of the requirements for, the role of, and the way to provision of what was the Cinderella of IT – Infrastructure. Get your approach right and you might be able to combine cost reduction, corporate green policy and still come out with the ability to support a wave of new business use.