"May you live in interesting times” is reputedly an ancient Chinese curse suggesting that the stress of chaos and change does not make for a good life."
There must be more than a few IT colleagues who feel they are indeed living in ‘interesting times’ at the moment. But until recently the key initiators of change were the business managers and users who they could at least try to screen out and keep on the edge of the business. Now it seems that the recent announcements by long established technology building block providers show they are set on contributing to the change too.
IBM’s new ‘PureSystems’ approach
IBM, who as a rule doesn’t do ‘superlative’ marketing statements, introduced their new approach to how hardware and software will be delivered in optimized fully integrated systems. Called ‘PureSystems’, this is “one of their most significant announcements in twenty years”. With a completely dedicated website it starts with the overview “expert integrated systems fundamentally change the experience and economics of IT with built in expertise, integration by design and a simplified experience”. Note the words fundamentally change and in this case it’s true for IBM. For the first time, the hardware and the software business units offer the results of their integration into a single unit to build and ship fully integrated products.
There is no doubt that the results will perform better, be simpler to manage, easier to troubleshoot. As IBM points out, there is just one dedicated support function and any arguments about faults on software or hardware issues will disappear. But it still means shifting to the mentality of ‘technology stacks’ and accepting that all the products will come from a single vendor. Maybe not much of an issue in a ‘forklift’ replacement or upgrade program, but that’s not likely to cover everything, so parallel operation of a mixed environment looks likely.
Oracle offers the ‘red’ stack
It’s the same story elsewhere with Oracle offering the ‘red’ stack. They set out the reasoning for its approach and benefits in a whitepaper which although a couple of years old still makes a good briefing on the topic. More recently the benefits and the push for using a full stack seem to have been much more around the Oracle eCommerce and web initiatives. And that again comes back to what the most likely approach most IT departments will adopt: the ‘a stack’ approach for delivery of their new ‘services, ‘web’ ‘cloud’ initiatives. Three terms that are becoming horrible confused and frequently interchangeable, but all share the common requirement for a virtualized online set of technologies.
HP talks of ‘converged’ as its approach to stacks
If IBM has a blue stack and Oracle a red stack then I am not quite sure what color the HP approach should be called. Possibly they would be happy to be thought of as a ‘transparent’ stack. I say this because HP are deliberately not trying to build the level of their stack to reach as ‘high up’ as their competitors in order to make themselves the natural partner for the pure software vendors. The terminology is also different with HP talking of ‘converged’ as its approach for clouds, data centers, and even networks. They acknowledge the need to bring old and new, both in terms of installed bases, strategic choices that need to continue to be used, as well as building to launch radically different requirements for ‘services’ on clouds. The question is, does this turn out to be a compromise in actual performance and other metrics against the more closed and highly optimized stacks?
Opportunities and challenges for IT departments
The reality for most IT departments is that for new non-traditional IT requirements, it will be simple enough to gain the advantages of a ‘green field’ approach and go for a stack to reduce the work of deployment and operations. The challenge is whether it makes sense to replace or integrate existing applications that are already in place and running. While every IT department will be different, here is an interesting write up and a good representation of what this might bring around when considering how to approach a JDE deployment. There is one other possibility and I know of an insurance company board currently asking itself why it would scrap its whole IT approach (it has become the usual expensive entanglement of customization with huge maintenance costs and use Commercial Off The Shelf, COTS) and replace it all on a non-capital investment basis.
Now that really is ‘fundamental change’ that qualifies as living in interesting times!