Information Technology Reference
In-Depth Information
workflows that represent “best practices” for scientific experimentation
can also be wrapped as services and published for others to use. In this
way, users not only consume but also contribute to the system.
This chapter starts with an introduction to the data deluge phe-
nomenon in science, and specifically in biological science. Then, we
present the paradigm of service-oriented science (SOS) in which Web
services are virtual access points to data and computational resources.
We present Cancer Grid (caGrid) as a case study of an SOS platform,
which has been developed under the Cancer Biomedical Informatics
Grid (caBIG) program sponsored by the US National Cancer Institute.
In Section 7.2, we discuss the requirements for developing scientific
workflows for caGrid and explain how we fulfill these requirements by
leveraging and enhancing existing software tools. The caGridWorkflow
Toolkit, an extension to the Taverna workflow system, is designed and
implemented to ease building and running caGrid workflows. Besides
the overall design and technical details of caGrid Workflow Toolkit, we
also present some real-life and service-based scientific workflows at the
end of this chapter.
7.1 SERVICE-ORIENTED INFRASTRUCTURE
FOR SCIENCE
7.1.1 Service-Oriented Scientific Exploration
The Fourth Paradigm
Scientific investigation has entered the age of
data-intensive science
,or
e-Science
. In January 2007, Jim Gray summarized the four science
paradigms chronologically, namely empirical, theoretical, computa-
tional, and data-intensive (e-Science). Here, we briefly explain each
paradigm by giving some examples. More information can be found in
the topic [74].
Empirical.
Galileo's observational astronomy that leads to the
telescopic confirmation of the phases of Venus, and the discovery
of the four largest satellites of Jupiter (1500s-1600s).
Theoretical.
Kepler's laws of planetary motion (1600s) and
Newton's laws of Motion (late 1600s) are representative.
