Cloud Computing and Enterprise Architecture
Here is a popular, new question that is posed to me: "What kind of impact is the rise in cloud computing having on enterprise architecture?"
Here is my response:
The whole idea of Enterprise Architecture is to enable the Enterprise to address orders of magnitude increases in complexity and orders of magnitude increases in the rate of change.
Therefore, if you have Enterprise Architecture, and if you have made that Enterprise Architecture explicit, and if you have designed it correctly, you should be able to change the Enterprise and/or its formalisms (that is, its systems, manual or automated) with minimum time, minimum disruption, and minimum cost.
I doubt that you would want to change the manual, material-handling aspects of your Enterprise based on Cloud Computing, and you may not want to change your strategic or core systems; however, if you wanted to change any automated systems, and if you have Enterprise Architecture designed for accommodating complexity and change, then you should be able to:
- Move the enterprise records off site to run on different data storage technologies and/or different database management systems products (Column 1: Rows 4, 5),
- Redefine the transaction flow based on different technology characteristics and/or recompile your instructions with different compiler products (Column 2: Rows 4, 5),
- Restructure the data network and address structures for storing your records and programs external to the Enterprise, running them on different operating systems (Column 3: Rows 4, 5),
- Reallocate the responsibility for producing the work products (screen formats and reports) to an agency outside of your jurisdictional control possibly using different formatting products (Column 4: Rows 4, 5),
- Redefine the control structures based on the different technological constraints and integrate those changes into the programmatic instructions (Column 5: Rows 4, 5),
- Change your technology motivations (your technology objectives and strategies) based on the Cloud infrastructure to identify the associated risks and modify your Business Rules appropriately (Column 6: Rows 4, 5)…
…all with minimum time, disruption, and cost.
In fact, if you have your Enterprise Architecture stored in a repository, I think I would SIMULATE those changes before I actually made any of them to make sure I did not introduce any discontinuities or dis-functionality that would disrupt the Enterprise business.
In fact, if you have your Enterprise Architecture and if the Repository tool in which it is maintained enables you to do these kinds of simulations, then I would also simulate how these potential changes could affect your Business Strategies (Row 1), your Business Concepts (Row 2), and your System Logic (Row 3) including the strategic systems and core systems to see if they would be impacted or if you could capitalize on these technology changes to re-think the Strategies (Row 1), the Business Concepts (Row 2), and the System Logic (Row 3) for business advantage. Then you could make modifications and re-transform those modifications into a new instantiation with new technologies (Row 4) and new technology products (Row 5) that you would have a high degree of assurance that it would work as expected.
While you are doing these simulations, you probably would want to examine the costs, disruption, and time for making the changes as well as the risks, especially if you are going to move things outside of your jurisdictional control.
ON THE OTHER HAND, if you DON'T have Enterprise Architecture made explicit and maintained or if it has NOT been designed to accommodate extreme complexity and extreme change, then I would start working fast and hard to create your Enterprise Architecture and engineer it to accommodate extreme complexity and extreme rates of change because Cloud Computing is not likely to be the last technology innovation you are going to have to address.
That's all well and good, but if you have to start from scratch and create your Enterprise Architecture, you are not likely to have it completed during this budget cycle… it may take many budget cycles to simply accumulate a critical mass of Enterprise Architecture to get you ready for the next change.
At some point, my opinion is, someone better start working on this because the opportunities for changes are going to get more complex and more invasive and come at an ever-accelerating pace… and not only technology innovations but environmental shocks, regulatory proclamations, and global competition. If you cannot assimilate the forthcoming changes, you are destined to go into "Future Shock"….
So… what do you do about Cloud Computing right now if you DON'T have Enterprise Architecture or if it is NOT designed to accommodate extreme complexity and extreme change? You have three options:
- Move whatever you want out to a Cloud and see if it works… maybe, after you get it moved, you might want to evaluate the cost, disruption, and time to move it and to operate it as well as any risks you have incurred by moving it outside of your jurisdictional control.
- If that is too high risk, then I would look at your inventory of data records, programmatic instructions, network structures and addresses, your screen formats and document formats, your technology timing cycles, and your technical objectives and strategies and compare them with the proposed new technology environment and develop a migration plan.
- If you don't have those inventories and it would take too long and cost too much to create them, then you can build some new portion of your Enterprise to run on some other technology environment on someone else's infrastructure. If you chose this option, the question will rapidly become, how does this new Enterprise implementation relate to the existing Enterprise legacy implementations and how can you resolve any conflicts and discontinuities?
So, in short, the question, "What kind of impact does the rise in Cloud Computing have on Enterprise Architecture?" is probably not the right way to ask the question. Probably a better way to ask it would be, "Is Cloud Computing a useful technology alternative for implementing our Enterprise (or at least portions of our Enterprise) and how would we affect that implementation to minimize our risk and maximize its utility?"
Then, I would suggest Enterprise Architecture is fundamental to answering this more appropriate question. In fact, I don't think you can answer the question without Enterprise Architecture. The only option you have is number 3 above, just build a new Cloud Computing implementation — in which case,
- you are adding one more implementation to the legacy of systems and likely recreating something that is already implemented (spending money that doesn't have to be spent) and
- adding general and administrative costs, additional on-going operating expenses to the Enterprise reconciling (interfacing) the new implementations to existing systems (less the cost/benefit realization of the new implementation) and
- increasing the existing complexity of business operations and
- inhibiting subsequent changes.
You can take refuge in the fact that the global playing field as it relates to Enterprise Architecture in 2011 is pretty flat… no one has substantial advantage. The question this begs however is, how long will that last???
Practically, in the short term, I would try out Cloud computing and try to understand its utility and its risks… but at the same time, I would start making the long-term investment in Enterprise Architecture. Enterprise Architecture, the knowledgebase of the Enterprise, creates the Knowledge Advantage paramount to Enterprise viability and success in the Information Age.
Furthermore… and even more importantly, Enterprise Architecture also facilitates the assimilation of extreme change, another even more daunting characteristic of the Information Age. The biggest challenge facing the Information Age Enterprise may well be change, not only because of its increasing frequency, but because of the unpredictability of its sources including technological change, environmental change, economic change, regulatory change, market change, political change, cultural change… etc.
Seven-thousand years of known history of humankind establishes that the only known strategy for accommodating extreme complexity and extreme change is…
This article can also be viewed on John's website — presented here, with permission.References
# # #