I saw 'time series database' in some marketing slides; the only bare definition I could glean from the 'net was 'A database that contains data for each point in time.'
The blurb on "Temporal Data and the Relational Model"
particularly on things like "a new approach to the problem of "granularity," is spot on.
Temporal database systems are systems that provide special support for storing, querying, and updating historical and/or future data. Current DBMSs provide essentially no temporal features at all, but this situation is likely to change soon for a variety of reasons; in fact, temporal databases are virtually certain to become important sooner rather than later, in the commercial world as well as in academia. This book provides an in-depth description of the foundations and principles on which those temporal DBMSs will be built. These foundations and principles are firmly rooted in the relational model of data; thus, they represent an evolutionary step, not a revolutionary one, and they will stand the test of time.
This book is arranged in three parts and a set of appendixes:
Preliminaries: Provides a detailed review of the relational model, and an overview of the Tutorial D language.
Laying the Foundations: Explains basic temporal data problems and introduces fundamental constructs and operators for addressing those problems.
Building on the Foundations: Applies the material of the previous part to issues of temporal database design, temporal constraints, temporal query and update, and much more.
Appendixes: Include annotated references and bibliography, implementation considerations, and other topics.
Describes a truly relational approach to the temporal data problem.
Addresses implementation as well as model issues.
Covers recent research on new database design techniques, a new normal form, new relational operators, new update operators, a new approach to the problem of "granularity," support for "cyclic point types," and other matters.