Scalability of the VSO

The successful development of the VSO prototype raises fundamental system architecture issues in dealing with the scalability and fault tolerance of distributed systems. The four sites participating in development (Stanford, MSU, NSO and SDAC) are only the starting point of VSO. The system’s overall architecture is being influenced by interaction issues between different data repositories that do not share a common set of capabilities. As new data providers or new search capabilities are added, the time needed to return useful results should not should not grow linearly with them. We take care to identify and isolate slow or unresponsive data repositories, to keep them from adversely affecting the overall quality of VSO searches. This extra complexity needs to remain unobtrusive to the solar researchers, so that they can focus on their research, and not the tools that they are using. Citation: Hourcle, J.A., Dimitoglou, G., Suarez-Sola, I., Tian, K., Davey, A.R. 2004, Scalability of the VSO, Bulletin of the American Astronomical Society, Vol. 36, p.796