What is Business Intelligence (BI), and how can the implementation of BI tools contribute to the success of an organization? Business Intelligence signifies the use of a variation of software applications and methodologies that enable organizations to collect data from internal applications and external data sources, to manipulate raw data, so that it can be evaluated for more educated business decisions by executives, managers and staff. BI assortment of tools, prepares data for analysis, create and execute queries against the data, and build reports and dashboard to make the analytic outcomes accessible to organization decision makers as well as managers and staff. Business Intelligence can incorporate historical data as well as current data. This allows for the prediction of trends and strategic and proactive decision making. BI tools were most found in the hands of data analysts, but now a days due the BI software becoming easier to use, and data discovery tools readily available, more and more business executive and staff member are able to use these tools. Allowing for data to be readily available and in real time.
As the demand increases for the need for BI teams, the size of the BI teams increase as well, the need for incorporating BI architect, developers, Business Analysts and data management personnel as well as business units. As these BI development process begin to grow the waterfall development approach is being replaced with Agile BI and data warehouse approach breaking the BI projects into small chunks and outcomes being delivered to the business in increments. This allows for speed of implementation and faster modification as the business needs change.
Difficulties that can arise from Business Intelligence is the fact that data being reported is not accurate. In other words data of poor quality. The poor data quality mostly comes from transactional application. (For example: Data is being enter wrong). This can cause inaccuracies in reporting leaving import decision doomed to fail, so it is import to solve the root cause of the problem. A good BI program will be able to uncover the holes early on, and saving the organization unnecessary expenses and poor decision making. At this point it is important to include data governance and data stewardship within the BI program, to oversee the data quality being reported on. It is important to make sure that the BI application architecture can support the future state of organization, and be easy to learn. If not BI is sure to fail. With that being said there is also need to have an underlining infrastructure stack to support the data warehouse, what I mean by that is, if you don’t have the hardware that can server up the data quickly and easily, the staff, business units and executives are going to look elsewhere to find their data.
With the demand for data analysis, organization must realize it is time to reach for Business Intelligence to transform data into information. Furthermore, in order to be competitive it is important to know how the organizations products and services are running. BI will be able to predict trends in an organization data, allowing for exploitation of holes and issues, and quick remediation to be taken, making sure the organization does not fail. Allowing the company to stay ahead of the competition, and increasing ROI and reducing risk to boot.

If we survey a group of people, we would more and likely receive a different answers on their perspective of the meaning to Service-Oriented Architecture (SOA). Reason being organization have different requirements and intentions for a SOA, because each organizations have different business markets and strategies. The basic definition of a Service-Oriented Architecture is simplified based on architecture designed to meet the business needs of an organization. The article refers to it as building software assets in the company using service-oriented programming methodology. Overall the origin of the SOA concept came from CORBA, DCOM, and DCE just to name a few, but what makes SOA improved is that it incorporates agile business process. Having agile co-exist with SOA, it bonds both execution and a sound architecture that will bring ROI to the business that the architecture alone could not achieve.
The term Web service describes a standardized way of integrating Web-based applications using the XML, SOAP, and WSDL. Web Services are not like the standard web page, because it does not utilize GUI interfaces for the user. Web services instead share business logic, data and processes through a programmatic interface across a network. The applications interfaces, not the users. It is the developer that adds the GUI to the web service. That being said SOAs may be understood via Web services but Web services are not needed when implementing SOAs.
SOA advantages, allows an organization to do first and for most is the use of repetitive software. This allow for quicker development in the future, by allowing a connection from new or existing systems to a centralized SOA system. When utilizing SOA it is important to have governance, so that centralization development, code repository, and one central agreed upon methodology is set in place. Another advantage utilizing SOA will increase productivity, by reducing work load for developers and increase turnaround time. The idea of connecting to pre-existing software will reduce cost, increase productivity, and produce a better product. An organization that is suitable for this type of development environment would be foolish not put SOA into practice.

Developing a well-defined Enterprise Application Architecture (EAA) can be tricky, especially when there is no buy-in from the business units and yes even some executives. Let’s start with the definition of enterprise application architecture, which is the description of the structure of the application and software use across the organization. The systems are broken down into sub system and connected to each other based on the relationship between them. Based on the article Application Architecture also verifies that the application is not only what the business needs, but if the application is scalable, reliable, available and manageable. The article continue to state the EAA not only deals with implementation, but lends a helping hand with risk management for technology and deployment strategy.
The basis of all good EAA is business requirements. This is why I believe it gets tricky at least within organization that rely on third party software, and allow purchasing to be left up to business units, and no interjection from IT. I am not saying that IT dictates what is needed, but they need to play a part. In my organization there are times IT plays second fiddle to the purchasing of software or even the building of it. In turn what happens 9 times out of 10 the vendor’s sales people promise the world, leaving out incompatibilities, infrastructure falls short, all because IT is not in the design phase. What usually happens is IT is blamed for the short comings. It is important for business and IT to co-exist for an EAA design to work successfully. This becoming ever so apparent with the rise of loud solutions becoming more prominent and software such as word press and expression engine being more readily available, it makes building a solid EAA even harder to do.
What we have done to combat these issues was first and foremost build trust in the business units. Then build a service portfolio, building out a DevOps team, as we are currently slowly building out an SOA. This SOA is allowing us to better interact with developers by utilizing such tools a GIT repositories, GITLAB, Puppet and Jenkins. These tools allow the developer to push code easily from the repository to the respective web sites. Allowing the code to go to QA easier and faster, which means shorter path to production. Since everything is automated it frees up IT time to work on new projects for the business. Leaving less stress all around and everyone is happier. Lessons learned is that performance will suffer at all levels if there is not a solid Enterprise Application Architecture and solid alignment from all involved.

2 thoughts on “Topic 2-Application Architecture

  1. I think it’s definitely important in the BI space as well as the data space in general to not focus on just the quality of data but the consistency. We had multiple instances in a previous organization where we took an approach of “garbage-in / garbage-out”. While on the surface that might seem odd the primary drive was to ensure that when a question was asked at the warehouse the same answer would result as when the question was asked at a data mart. In highly regulated environments this can often be as important as getting the “right” answer. Interesting post. — William Gates (EA874)

  2. Nice post on BI. A few comments on live data and input/output of bad data. We run SAP HANA and although I’m not directly involved with it I did ask around when I started this class and the app provides near real time data, and like most executive like to see, colorful charts! It’s also a matter of devices too, iPads (which I’m typing on now) is typically used by executives or sales and is a must have reporting tool. If it doesn’t display on the iPad then its not the right app.

    As far as data in/out, you being up a great point and a very relevant one. If the data coming in isn’t cleaned properly it will only output invalid data. Obviously realtime data that isn’t being presented right can be catastrophic to executive or field sales. For sales in can make or break a deal. A very important factor that often gets assumed to be right, data cleansing is a vital role in the proper output.

    Thanks for your post!

Leave a Reply

Your email address will not be published. Required fields are marked *