Enhancing Data Sharing in Collaborative Research Projects with DASH
Tom Ferrin
Departments of Phamraceutical Chemistry and Biopharmaceutical Sciences
University of California, San Francisco

DASH (DAta SHaring) is an infrastructure for facilitating the
construction, maintenance, and dissemination of shared data among
researchers in small-to-mid sized collaborative projects.  The DASH
model is built around a "data network" comprised of data sources
(database tables, filesystem files), and protocols (executable
programs) that use those data sources as input and/or output.
DASH monitors your data, detects changes as they occur, and responds
to those changes in a timely manner. DASH can be used as a workflow
management system for projects such as scientific databases or other
large scale data-centric applications.

While in traditional workflow management systems, a set of workflows
are defined, then executed manually, DASH lets you define a complex
web of workflows, and then decides which of these workflows need
to be run in response to new or modified data in the system.  While
DASH runs on a single machine, it exposes a network API that enables
DASH instances running on other machines to initiate workflows on
the local instance, or receive notification of any changes to the
local instances monitored data.

The set of tools provided by DASH is both comprehensive enough to
serve as the infrastructure for new collaborative projects, and
flexible enough to be applied to existing ones.