Tuesday, May 3, 2016

Relational Database Management Systems

According to Coronel, Morris and Rob, databases are shared and integrated computer structures that store a collection of end-user data and metadata in a methodical way that is much more dynamic, malleable and synchronized than the traditional file system.  Databases avoid some flaws of the file system, for example, it's possible to add single bits of unconnected data without the requirement of adding filler data to a ledger's expansive columns and, secondly, deletion/edit anomalies no longer occur.  It's easy to understand how databases could be valuable when a company has trillions of pieces of raw data to manage, process and store.  Also, databases allow for quick transformations of raw data into graphical or tabular presentations of more meaningful information, which improve business decision-making.

I have no prior education nor experience in database modeling, i.e., representing and storing data, nor management software, which is used to create/generate, manage, store and retrieve/query the interrelated data by the single or multi-user end user(s), so I'm looking forward to learning it, other database concepts and statement writing, and, in particular, MySQL.  According to Coronel et al., databases result in improved data sharing, security, integration, productivity and access, as well as minimized inconsistency.  Single users have desktop databases and multi-users either workspace or enterprise databases.  Also, if the data is located at a single site, it's centralized, in contrast to distributed.

General purpose databases are varied and discipline specific ones will be more specific to its related discipline, such as medical of financial records.  Alternatively, operational databases may focus on something like a transaction occurring day-to-day.  Analytical databases focus on storing metrics used for tactical decision making, having "massaged" or manipulated raw data to extract valuable information.  This data is made structured by the processing of raw data to business intelligence.  Most databases mentioned above will use at least semi-structured data, perhaps that made textual by XML.

The specific database model we will study is the predominant relational one, which has the logical structuring of related tables, connected by super, primary, candidate, foreign and other keys, central to its system.

After listening to class orientation, my understanding is that we'll largely be studying relational database models and primarily using Oracle, in particular, with a focus on foundational theory and design, so I'll be interested to compare the two software and study implementing the former, i.e., MySQL, on my own, if necessary, given its ubiquity in the workplace.  (I'm hopeful the divide between theory and practice won't be too great, in this respect, as is sometimes the case in academia, and that I'll develop not only a foundation here but a working, transferable knowledge that will be applicable in a professional capacity such as database developer, designer, administrator, architect, consultant, etc..)  My aim is to gain a working proficiency that includes implementation of these, and I'm looking forward to this class!   Still, I appreciate the value of learning strong design fundamentals, which is crucial.

I had a slight delay getting started this week, as I went to the the south bay to collect a PC from my sister, who's agreed to let me borrow it for the length of the class.  There were no issues installing Oracle, which I left overnight and returned to in the morning.

Reference

Coronel, Carlos, Steven Morris, Peter Rob, and Carlos Coronel. Database Principles: Fundamentals of Design, Implementation, and Management. Vol. 10. Mason, OH: South-Western, 2013. Print.

No comments:

Post a Comment