JPL Technical Report Server

Key ingredients needed when building large data processing systems for scientists

Show simple item record

dc.contributor.author Miller, K. C. en_US
dc.date.accessioned 2004-09-17T19:42:26Z
dc.date.available 2004-09-17T19:42:26Z
dc.date.issued 2002
dc.identifier.citation IT Symposium en_US
dc.identifier.citation Pasadena, CA, USA en_US
dc.identifier.clearanceno 02-2802
dc.identifier.uri http://hdl.handle.net/2014/10942
dc.description.abstract Why is building a large science software system so painful? Weren't teams of software engineers supposed to make life easier for scientists? Does it sometimes feel as if it would be easier to write the million lines of code in Fortran 77 yourself? The cause of this dissatisfaction is that many of the needs of the science customer remain hidden in discussions with software engineers until after a system has already been built. In fact, many of the hidden needs of the science customer conflict with stated needs and are therefore very difficult to meet unless they are addressed from the outset in a system's architectural requirements. What's missing is the consideration of a small set of key software properties in initial agreements about the requirements, the design and the cost of the system. en_US
dc.format.extent 3297033 bytes
dc.format.mimetype application/pdf
dc.language.iso en_US
dc.subject.other Earth Observing System EOS en_US
dc.title Key ingredients needed when building large data processing systems for scientists en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search


Browse

My Account