Database systems—breaking out of the box
Avi Silberschatz, Stan Zdonik · ACM SIGMOD Record · 1997
This report takes a somewhat different tack. Our theme is that database research should be devoted to the problems of data management no matter where and in what form the data might be found. We should not be defined strictly by the current product space or by the commonly held notion that our job is to manage very large collections of structured records within a controlled environment. Instead, we should apply our skills to new data management environments that potentially require radically new software architectures. 2 Background The database field was born in the late 60's with the release of IMS, an IBM product that managed data as hierarchies. While hierarchies later proved to be too restrictive, the key contribution of IMS was the widespead revelation that data has value and should be managed independently of any single application. Previously, applications owned private data files that often duplicated data from other files. With a DBMS, data should not be logically replicated, making it easier to maintain. Creating shared databases required analysis and design that balanced the needs of multiple applications, thereby improving the overall management of data resources Both the IMS data model and its most well-known successor, CODASYL, were based on graphbased data structures. While the idea of traversing links was intuitively attractive, it made it difficult to express database interactions independently of the actual algorithms that were need to implement them. In 1970, Ted Codd published a landmark paper [Codd70] that suggested that data could be managed at a much higher level by conceptualizing it in terms of mathematical relations. This paper sparked a great deal of interest within the research community thoughout the 1970's to make this notion practical. A...