This is an application from the Center for Clinical Epidemiology and Biostatistics (CCEB) at the University of Pennsylvania (Penn) School of Medicine to upgrade and expand the research computing infrastructure, in support of a multiple, meritorious NIH-funded research projects. This proposal requests a SUN Microsystems SUNFIRE 6800, a specialized database and statistical compute server, to provide new capabilities, capacity and reliability, by replacing a dated Oracle (r) database and inadequate statistical computing infrastructure. This server is particularly well suited for supporting large scale database and statistical compute-intensive environments. This application advances the following specific aims: 1) replace, and upgrade the current database computing environment, thereby ensuring continued reliable support of active NIH projects., 2) provide a responsive statistical analysis computing environment for the NIH projects currently relying upon the database environment, and 3) provide improved computing performance and expanded capacity so that additional NIH projects can be supported. The Clinical Research Computing Unit (CRCU), a School of Medicine Core Research Facility within the CCEB, supports the research computing infrastructure required by large-scale clinical research projects, and statistical methodology grants. Currently supported projects (30 active NIH funded projects with total costs of $190,235,076) range from statistical methodology ROls, to clinical research RO1 and PO1 projects, to cooperative agreements (U01, U10, U18) that include multi-center, multi-study, Data Coordinating Centers (DCCs) and Center for Education and Research Therapeutics (CERTs). These NIH-funded research projects rely heavily on advanced database technologies, with many utilizing state-of-the-art WWW-deployed data management systems and databases for remote data entry and management. All projects involve extensive statistical analysis, with a growing need for database and statistical compute-intensive infrastructure. To provide an adequate computing environment, the CRCU maintains specialized computing equipment specifically configured for the Oracle (r) Relational Database Management System, SAS (r), and S-PLUS (r), MATLAB (r), and STATA (r). However, the hardware requirements to support the expanding NIH research program at Penn far exceed the limited resources provided by standard compute servers, originally installed in 1997, and are very consistent with the equipment eligibility criteria set forth in the RFA. Growing demand for NIH project support, and the advanced age of the original computing infrastructure, is causing capacity (i.e., performance and number of supportable projects) and dependability limits to be reached and exceeded. Each additional NIH project further stresses the reliability and stability of the original equipment, and provides only marginal funding for upgrading the original computing infrastructure. This large computing infrastructure request is highly consistent with NCRR goals of providing shared resources across multiple grants that are cost-prohibitive and difficult to justify for individual projects.