Hi Folks!

For a while I was attempting (probably the correct word) to keep both this blog as well as my SQL Server Magazine SQL Server BI blog current. Working on growing B.I. Voyage , speaking, writing, and blogging is quite the work load. And so this blog suffered as a result. THANK YOU all who read my blog here on SQLTeam.com, I intend to leave it live for those who might benefit from its content at a later date. My consolidated, single blog can be found at http://www.sqlmag.com/blogs/sqlserverbi.aspx .

Cheers,

Derek

As i look around at some of the leading companies across the globe, I find a state of chaos across these companies, especially in information management. What is interesting to note is that the problem transcends Fortune 100 to Small Business and is very similar. Where did things go wrong? all these companies have spent valuable dollars to improve information management.

Deeper investigation reveals several gaps and loopholes. The vendors that came to provide strategy and technology services addressed “point in time” problems and developed focused solutions. With the craziness of technology improvements, market conditions (m&a), business changes all coming together, the entire situation is like playing “poker” at the high stakes tables in Vegas.

How do we unravel from the state of chaos to a state of order. Hiring more strategy consultants is not the only answer. We need to do a holistic overview of the business situation and find out where issues have come from and how the current solution process will address today and the future.

In order to recover from a state of chaos to clarity, you need to look at the strength of the organization and its people, assess its technological prowess, its competition, business value, market value, strategic maturity and adoption of processes in the organization and more.

However doing all this in a short span requires a great deal of focus and more maturity from the vendor that will be hired for this exercise. Size does not matter in this spectrum.

A vendor rating from the Analyst community is a starting edge of this process, but that alone is not enough, we need to ask for organization maturity, delivery capability and much more to not repeat the mistakes to get to a state of chaos. Sometimes a few “niche” vendors may be needed to solve the larger problem.

To be continued…..

We all know by now the word “columnar database” and some of the most popular providers – ParAccel, Vertica and Infobright. Though columnar databases have distinct performance advantages and applications that can be deployed, from a business perspective, there was a question mark and doubts of where to apply these solutions.

In my opinion, the unstructured data integration into the data warehouse is a very key area of applicability for the Columnar database. The reason for this statement, stems from the fact that the unstructured database is a very column oriented data store and if we are processing oodles of text, the reference of information will be more clustered in nature. This drives the need to store the data in a co-located manner, which then leads me to look at columnar databases more closely.

I do not imply that Oracle, SQL Server, DB2 or Teradata cannot support the unstructured database needs, but for those who want to adopt to the cloud, use more on demand scalability etc, columnar databases may be an in-house option. Additionally for those providers in cloud arena, columnar databases may be more optimal to adopt to.

I’m currently running tests on Infobright and will share the results in the next few days. I’m processing a large volume of semi-structured and unstructured data and will measure the throughput and performance against the standard RDBMS platforms.

In my technical opinion, choosing a columnar database for processing large textual data provides more business benefits, the final results can be migrated to a corporate environment, but operational aspects can be done in the columnar platform.

I have been researching on this subject for the past few months. It is becoming clear that to control errors in diagnosis and treatment of medical conditions, Evidence Based Medicine can be defined as “the conscientious, explicit
and judicious use of current best evidence in making
decisions about the care of the individual patient.
It means integrating individual clinical expertise
with the best available external clinical evidence
from systematic research.” (Sackett D, 1996).

This requires research and processing of clinical information that can be leveraged by a Doctor. While search from Google, Oracle and others are supporting the search portions, Clinical data is heavily semi-structured – a combination of structured and unstructured data. The unstructured portions is where you have all the content and context of diagnosis, treatments, prognosis, conclusions etc. This is where Textual ETL and the unstructured database build will help us. The entire DW2.0 methodology can be directly applied to this system. We can classify, categorize, contextualize and capture information from the clinical data into a database.

In conducting the research, I have found that we can process years of clinical data into an unstructured database with relative ease and when combined with the statistical research data, the results have been an eye-opener. The possibilities of reducing errors and offering the right treatment means the cost of healthcare can be managed well. The quality of the treatment will be more rewarding and welcome.

I know that many Doctors and other medical professionals have different opinions on this subject, but that was when no technology was available to support EBM, now that we do, we should start paying attention to this area.

I will be writing my first article for this year on this subject, stay tuned for the article.

I’m stumped by irrational behavior from large technology corporations, especially ones that feel that they have big muscles and have earned the right to such behavior. They go to any lengths to be winning, and do not realize that what they think is collateral damage from their perspective, can be a serious business factor in the future.

What do these companies think? do they even think before they get on an assault charade? do they care for ethics and values? if they do not, do we also start questioning their ethics and values?

The end result of such negative behavior is the loss of trust in the long course from their own customers and marketplace. Sadly their focus is on today’s gain and not the overall future perspective probably, which to an extent explains such behaviors. One does not know how many times these companies have lost business due to these behaviors.

I hope that at some point in time, these corporations and their management do realize the value of earning the trust of their customers and the market is very important for their continued success and it is not just technology or financial glories that carry an impact.