SBIR-STTR Award

Today’s high energy physics research demands systems that allow collaborators to efficiently analyze vast sums of data, while using limited computing resources. The data intensive requirements of particle and nuclear physics combined with the distributed nature of both computing resources and physicists make this a unique problem. At Fermilab, the SAM System serves the distributed data access needs of collaborators throughout the world, providing physicists with an easy to use interface and masking the complexities of the underlying meta-data storage structures. This project will extend the robustness and capabilities of the SAM Query Language to provide a solution adaptable to other particle physics facilities, while deploying implementations pertinent to additional problems at Fermilab. In Phase I, additional infrastructure enhancements will be added to provide an accurately documented and implemented grammar syntax checker, to enforce security authentication, and to ensure maximal database resource utilization via performance optimization and efficiency techniques. Next, interface extensions will be added to ensure the query language is easy to use and adaptable to other particle physics data access problems. Finally, a detailed analysis will be performed of the needs of the D0 Experiment for additional domain areas that need grid like meta-data query access to raw data.

Commercial Applications and Other Benefits as described by the awardee:
The enhancements to the SAM Query Language should be applicable to a myriad of data access problems facing other scientific institutions. As data storage and access needs grow in other traditional corporations and government agencies, they too may seek such alternative, cost effective means of storage and data access

As high energy and nuclear physics data needs grow, so too do the scale, size, complexity, and cost of solutions for storing and querying large databases. This project will develop technology that will allow users to submit a single query that can be routed to multiple, distributed, heterogeneous databases.  By combining the various databases into a “virtual data warehouse”, this Database Grid Solution will provide a single result set, masking the complexities required to find the actual data sources.  Phase I extended the design of an existing  query mechanism, the Sequential Data Access via Meta-data (SAM) Database Server, to increase the robustness and scalability of the current implicit query capabilities.  The new designs included more robust grammar parsing and syntax checking of queries, integration of Grid security mechanisms, database resource usage monitoring and limiting, and extended user interface capabilities.  With software leveraged from other sources, Phase II will integrate these designs into the Grid framework, ensuring that the end solution readily adapts to existing Grid standards and emerging Grid specifications.  A working example will be produced by adapting the technology to the current SAM Database Server  Commercial Applications and Other Benefits as described by awardee:  The Database Grid Solution could be applied to any ogranization that relies upon a complex, multi-database environment.  For example, large conglomerates attempting to leverage their existing worldwide data assets could do so without incurring the significant cost of combining them into one data source.  Local not-for-profit groups could pool their data resources, providing a stronger service for their user community.  Organizations involved in mergers and acquisitions need not spend a great deal of effort combining databases, but merely could deploy a virtually combined data warehouse.