The SEQ project
addressed the following fundamental
should we extend relational database systems to support queries over
ordered collections of records, rather than sets or multisets of
records? This work led to the SEQ system at Wisconsin, and later
to the Predator system at Cornell (which also built upon the Shore
storage manager), and focused attention on how to model order in a
relational system. The language proposal that resulted, SRQL,
extended SQL to
support "window" queries over tables ordered by sorting, and is
essentially the WINDOW function now included in the SQL:1999 standard.
This work anticipated much of the subsequent research on stream
databases, which deal fundamentally with ordered data and window
queries, but additionally address issues such as arrival and processing
rates and continuous queries.
S. Beyer, Muralidhar
SRQL: Sorted Relational Query Language. SSDBM
Probabilistic Optimization of Top N Queries.
The Design and Implementation of a Sequence Database System. VLDB
E-ADTs: Turbo-Charging Complex Data. IEEE
Data Eng. Bull. 19(4): 11-18(1996)
Management of Sequence Data.
Ph.D. thesis, Univ. of Wisconsin-Madison, CS Department 1996
SEQ: A Model for Sequence Databases. ICDE
Sequence Query Processing. SIGMOD
Conference 1994: 430-441