The assigned reading from Access Science gave a broad overview of databases, which was surely the intent. The article broke the idea of databases down into sections such as hardware, database architecture, language etc. While the article provided an introduction into the idea of relational database management systems, it was nothing ground breaking. I found the most interesting part of the article to be the emerging technologies and the future of databases within mobile networks and technology.
The article I read, Does Relational Database Perish in the Future, discussed the future of data storage and the role that relational databases may play. The article talks about how the majority of data today is stored on Relational Database Management Systems (RDBM), yet this is in many ways a dated storage technology in terms of storage and analysis. The article then went on to discuss how the decreased price of data storage, along with the ever increasing amount of data produced has resulted in a "data dump" attitude. To better sort and store the data accumulated, Google uses MAPREDUCE. MAPREDUCE "is a programming model and an associated implementation for processing and generating large datasets that is amenable to a broad variety of real-world tasks"(Dean and Ghemawat 2008). Google began using this system of clusters to storage and sort data gained from user's email and searches, allowing for optimized search engines and ads. At the end of the article, the author proposing that traditional relational databases will most likely be replaced by optimized database storage systems such as MAPREDUCE.
I also found the emerging technologies part very interesting. Im interested to see what the future holds for database systems in terms of their applicaions and uses.
ReplyDeleteOne of the articles I read was on the issue of huge amounts of data stored in relational databases as yours. It dates to 2005 so maybe that's why there wasn't anything about Google's solution but what the author proposed was using flat files such as text files along with RDBMS to increase their speed, decrease the amount of data stored and the cost. It seems in one way or another traditional relational databases need to be modified.
ReplyDelete