Completed Berkeley Database Projects
adaptive, multiuser continuous query engine for data streams and
remote data sources. It features an adaptive query processing
infrastructure built on Eddies and related technologies, supporting
shared processing of multiple overlapping queries, and extensible
handling of streams under overload. Additional research topics
include support for parallel fault tolerance and load balancing,
archive management and querying, and partial results to ongoing
queries. Telegraph has been used in a number of applications,
including high-speed network monitoring.
internet-scale query processor that builds upon peer-to-peer
Distributed Hash Table (DHT) technologies to enable SQL-style
and continuous queries over massively distributed data sources.
are XML message brokers
; systems that filter, transform,
and route XML documents on the fly.
embedded query processor for wireless sensor networks. It allows data
acquisition and query execution to be specified in a declarative
manner, and optimizes such specifications into distributed
computations over an ad hoc wireless network.
The CONTROL project studied interactive, online processing of large data sets.
It included work on Online Aggregation and query processing, online
association rule mining
online data visualization (CLOUDS
and online cleaning and transformation for large data sets
(Potter's Wheel ABC
The Potter's Wheel work was incorporated into the Cohera Catalog Manager,
now part of
Oracle's PeopleSoft Catalog Management
The GiST project studied engineering and mathematical fundamentals for
indexing of massive amounts of complex data. It included the libGiST
C++ library and the
analysis framework; an implementation of GiST indexes for PostgreSQL
that is now part of its standard distribution,
and a key piece of PostGIS
and research results on indexing and indexability.
A scientific database workflow system. Subprojects include:
the NASA Earth Observing System Data and Information System
Alternative Architecture prototype.
ESMDIS: the Earth System Model Data Information System prototype.
part of a
to create a federation of data producers.
BigSur has been commercialized by
Berkeley Earth Science Tools, Inc.
A database visualization system. The
designed and prototyped a database browser and visual programming language (using a
``boxes and arrows'' paradigm) for scientific applications.
The ideas behind DataSplash have been commercialized in the Visionary product of Informix Software, Inc.
Version 0.10 was the last public release of this software. As of 1998, this software is no longer being developed or supported by U. C.
An agoric distributed database system. The
designed and prototyped a scalable (10000+ site) wide-area distributed data manager.
Mariposa has been commercialized by Cohera Corp. (now a part of Oracle).
Version Alpha-1 was the last public release of Mariposa. As of 1997, this software is no longer being developed or supported by U. C.
A group of volunteers significantly enhanced University POSTGRES 4.2 (SQL support, working aggregates, etc.), resulting in
PostgreSQL (at one time known
The University POSTGRES project, which demonstrated the practicality of object-relational technology, was
discontinued in 1994.
POSTGRES was commercialized by Illustra Information Technologies,
which was purchased by Informix Corporation, which in turn was
purchased by IBM.
Version 4.2 was the last public release of University POSTGRES. As of 1994, this software is no longer being developed or supported by
U. C. Berkeley.
The University INGRES project, one of the pioneering relational database management systems, was discontinued in 1985.
INGRES was commercialized by Relational Technology, Inc. (a.k.a. Ingres Corp., a division of The ASK Group; subsequently part of Computer Associates, now Actian Corp.).
Version 8.9 was the last public release of University INGRES. As of 1985, this software is no longer being developed or supported by U.
C. Berkeley. However, commercial
was released as open source.