Research
Current Projects
- Digital Data Provenance
- In-Cloud metadata for eScience data (XMC Cat)
- LEAD
- XBaya
- Sigri
Detailed information about DSI's work in developing tools for provenance generation and collection and case-based reasoning. [+]
XMC Cat is a toolkit for capturing and storing metadata gathered during the scientific workflow execution. Its advantages include support for automatic creation metadata through curation plugins needed for subsequent discovery and use purposes. It provides search and browse capability through a client side toolkit and GUIs. XMCcat can be adapted to new scientific domains through configuration changes. That is, new code need not be written. It is currently in use in the LEAD Science Gateway. [+]
Linked Environments for Atmospheric Discovery (LEAD) is a collaboration between meteorologists, computer scientists, and educators to build a cyberinfrastructure that enables adaptive weather forecasting, forecasting that is responsive to and can focus on emerging local severe weather conditions. LEAD is funded through NSF.
XBaya is a graphical client program for workflow composition, execution, and monitoring. It allows the users to compose workflows from Web Services and wire them together in a interactive way to construct workflows. The composed workflow can be exported to various workflow languages. Primarily it supports BPEL which can be deployed into Apache ODE or GPEL workflow engines for execution. It also support Jython script based workflow execution and also very interactive Java enactor for workflow execution which enhances the user interaction aspect of the workflows. XBaya also allow the users to graphically monitor the execution of workflow asynchronously using WS-Eventing based publish/subscribe mechanism. [+]
e-Science applications are often compute and data intensive, requiring large-scale compute systems for execution. Large-scale systems, however, support a variety of resource management interfaces. Grid middleware solutions abstract these heterogeneous resource managers and offer a single unified job management interface. However, Grid middleware tends to be highly complex, needing technically sophisticated system administration skills to deploy and maintain these services. Further, many clusters in the academic setting are not part of a larger scale grid and have to be directly accessed by non-uniform vendor specific resource managers. With the goals of providing a simple, reliable and highly scalable uniform job management, we introduce Sigiri, a light-weight job management and abstraction service. Sigiri supports existing popular job specifications like JSDL and RSL. A Web Service Interface is provided to easily integrate with various scientific workflow systems and each step in job submission and management is decoupled to increase scalability. [+]
Archived Projects
- dQUOB: dynamic Query Objects
- Calder Complex Events Processing
- Relational Grid Resources (RGR) Project
- Doppler Source
dQUOB is a middleware system providing continuous evaluation of queries over time sequenced data through an SQL-like language interface. The dQUOB system has been applied to such diverse applications as a safety critical autonomous robotics simulation, and scientific software visualization for global atmospheric transport modeling. [+]
Calder researches complex events processing in data driven scientific computing. Current work investigates complex scientific use cases for events processing and programming models that integrate complex events processing and service oriented workflow systems. Earlier work investigated provenance for data streams. The work is funded through a DOE CAREER grant. [+]
The Relational Grid Resource project explored the representation of grid resource data. It advanced the thinking about resource information (e.g., CPU speeds, amount of memory, I/O bandwidth, storage resources) by showing that existing solutions using LDAP were inadequate for storing resource information because resource information underwent rapid change. This work was funded through an NSF ITR grant. [+]
The Doppler Source project processed streaming data from the NEXRAD radar network comprising 120 operational WSR-88D Doppler radars in the continental United States. It automatically generated metadata on the fly and stored the data to a SQLserver database. The files themselves were stored to a tape archive, then retrieved using attributes gathered during metadata generation. This work forms the basis for some of the ideas in the XMCcat project. This work was funded by Microsoft.
