NLM Home Page VHP Home Page

Next: References Up: Title Page Previous: Results Index: Full Text Index Contents: Conference Page 

Conclusions

      With the data obtained from this pilot project, it is now understood that several aspects of planning and development will need to be addressed in parallel: Evaluation of standard Internet and vBNS network performance (to be done in collaboration with the Pittsburgh Supercomputing Center and Carnegie Mellon University); Evaluation of Content penetration to medical, dental and allied health students will be performed using the UM Medical School Gross Anatomy ciriculum in a beta production environment; Enhanced Java software development; Algorithm, data fusion, and other advanced image processing tools will be needed; Label integration, and object-relational databasing, needs further specification; System and software scalability issues need to be determined.

      The greatest challenges preventing the full use of the datasets are their relatively large processor, memory, and internet bandwidth requirements. This has limited the number of qualified and motivated developers; even those few, like ourselves, with access to the finest of high performance computing (HPC) and networking environments.

      We are currently planning a complete system to (a) Greatly improve our quality of services offered by leveraging the NGI and its enhanced network management; and (b) enhance our capability to deliver service by developing a system and architecture to make use of the improved infrastructure technology for collaboration provided by NGI to develop, demonstrate, test, maintain, and evaluate a high speed distributed educational and research resource in the area of revolutionary medical applications for the NGI utilizing the VHP (NLM Award Number NO1-LM-8-5630)

      Figure 3: Hardware, software, and functional issues to be investigated to determine the appropriate enhancements and activities to ensure 7 x 24 delivery of the UM NGI VH product.

      This work was supported in part by the Advanced Research Projects Agency (ARPA), DAMD17-94-J-4512. We would like to thank SGI for their technical support and extended workstation loan.

Next: References Up: Title Page Previous: Results Index: Full Text Index Contents: Conference Page