Please login or join to subscribe to this thread
Decide what your System of Order is to find out where to start.
You have two approaches: architecture it out (top-down) or microservice it (work on multiple services/DBs, systems at the same time and connect them together)
Then find out if you have enough resources/people/knowledge internally or you have to outsource parts (or all).
Take waterfall approach to your planning and execution as it is best suited for infra projects.
Your biggest risks are huge cost overruns, schedule overruns, and problems with managing all stakeholders (and their unrealistic expectations).
Where do you start? Start like any other project with a business case then a project charter.
What do you upgrade first? Start with the riskiest or the most valuable infrastructure item. The former will expose the risk early enough for you to adjust accordingly. The latter will provide value to your users.
How do we capture total scope? Start by defining the scope boundaries - what is included, what is not included.
What are the typical risks involved? Same risks as any other project. Because infrastructure usually involves more suppliers, expect stakeholder risks to be particularly demanding.
Who are usually the major stakeholders? Answer: Same stakeholders as any other project, with a lot more suppliers of hardware, software, and services.
Before you make any attempt to undertake this investment in resources make sure you have performed an extensive business case and looked at all the alternatives to mitigate risk, time to complete project, cost to complete project and different IT platforms that exists.
These days large organizations are trying to avoid such capital outlays by either outsourcing their IT infrastructure to a managed services partner were the investment is shared and broken down over a number of financial years.
Using someone else's data centres and fiber backbone connectivity by partnering with an IT managed services partner is a far more realistic approach in order to achieve the same objectives and goals.
Also this would also fall under the remit of a program manager as this undertaking will consist of numerous separate but inter related project over a number of years so I would approach this undertaken in such a realistic manner.
Also the organizations Management Information Systems are the heart of the IT infrastructure so If you can find a solution for connectivity that is inexpensive, that is public cloud solution as apposed to leased line solution this will dramatically reduce your annual recurrent capital outlay.
Assuming that we are talking a large scale effort, i would recommend outsourcing it to someone who is in the business of doing such projects. The way i look at it, you need someone who already knows the dos and donts and can save you a lot of thinking while your organization can continue to do their day jobs.
Such a partner would typically help you come up with business case and scope and at that point you can decide how much you can do internally and what you can outsource.
I was involved on that three years ago including it we moving a holde data center from Mexico to USA. To take it from a perspective take into account that will be a solution for a business problem. Then, first thing to do is to work on the problem understanding that problem is the gap between the reality as perceived and the reality as desiere so you can work on perception, on desire or on the gap. When you have that, take into account the enterprise architecture to define the solution. Time ago I wrote an article that was published by the PMI and others as "best practice". Hope it helps: https://www.projectmanagement.com/blog-pos...-right-solution
Please login or join to reply