you don’t have to be a fan of the data vault to understand or appreciate the value in systems consolidation right? do you have 3, 5, 15 or more source systems that are flapping in the wind and not talking to each other? do you have disparate star schemas roaming different branches producing a myriad of results without a centeral data warehouse? do you have a master data initiative or need to start one? consolidation is in the fore-front of today’s it cost-savings, over the next set of blog entries i will address the issues, and see if i can provide some much needed guidance in this area. oh yea, for you who are data vault fans – thanks! you’ll see some value in this as well.
you’re probably a lot like me, have years of experience in the field, built hundreds of systems, seen it all and wear heavy flak jackets right? well, how many of you out there face day to day problems where systems need to be consolidated but it seems to be a nightmarish task?
in this blog, i’ll start by defining what i think are the issues, and because i won’t be able to get them all i hope you’ll chime in and respond by adding your own (the ones you run in to on your day to day activities).
first: what is consolidation? what are we talking about here?
well, to be quite frank – we are talking about systems consolidation – the idea that we can take a number of data sources or data targets that are disparate, and consolidate them – or merge them into a single storage system with singular maintenance cycles.
who might need consolidation?
- mergers and acquisitions – frequently need to “see what value the combined systems data would bring” before they decide to complete the merger. also, after acquisition, they need a corporate master view of the data set from both organizations in order to make intelligent executive decisions.
- departments who are retiring old systems – frequently need to merge several systems together, again for a single departmental master view of the data, so they can also load it to the new system appropriately
- executives who are looking at high maintenance costs and massive rentention policies along with upkeep of geographically disparate systems – they will need a consolidation of the information to a single system to be green. to save money, to reduce overhead, and to comply with audits and accountability.
- it staff who are engaged in a master data project – they will need to look at operational data sets in order to construct a consolidated view of the information in a single place.
- it staff who are engaged in an enterprise data warehouse project – they will need to look at consolidating disparate star schemas and a number of other systems that produce “reports” for different parts of the organization.
it’s no secret that consolidation is on the minds and lips of executive staff around the world. just look at the rise of out-sourced it solutions like cloud computing (not all it’s cracked up to be). there’s been a rise in consolidation of data centers as well to cut down on energy consumption and emissions. what it amounts to is the corporation who goes green, pays less in energy bills, less in maintenance, and has less overhead for the cooling systems and clean rooms in which the servers are stored.
so if consolidation is already under-way, what are we missing?
we (in it) may not be missing it, but in business it’s often over-looked: data consolidation and therefore data model architecture consolidation. shouldn’t we be looking at consolidating disparate data sets? shouldn’t we be looking at integrating back-end architectures? shouldn’t we be consolidating star-schemas, disparate data warehouses, and data junk-yards?
does it make sense to you, to attack a master data solution the same way we have been attacking “data delivery?” by that, i mean: for years, we’ve been building and deploying point solutions. yes it’s fast and low cost up-front, but look at what it costs in hindsight!! here we go again – building operational bi, and master data systems in a disparate and separate environment… one for each department, heck why not one for each user that can afford it? where’s the master data in that?
ok – enough ranting: what do you propose?
in this series of posts, i will propose an approach to consolidation – that if done properly, can and will save you money, time, and head-ache. along the way, i will introduce you to some ideas in a new breed of automation tools that can enable you to reach your goals better, faster, and cheaper. all with less errors!
now it’s your turn…
get involved, post your comments and replies on this blog – let me know what you run into day to day, and what you wish you had a tool to help you with… what’s your panacea?
in the next blog post i’ll dive in to each of the problems, and try to explain the value of consolidation.