>>> Is there a better solution to managing large quantities of data than SQL?
I am incapable of answering that, Zim.
I am not saying 'SQL' is bad for databasing the broad array of massive 'data sets' that it can handle.
Just that SQL is nasty *for me*.
Is it fast and efficient ? Depends on the type of data.
An SQL database for handling large 3D seismic data would not be nearly as fast nor efficient on disk storage space as even the most simplistic of 'seismic volumes' stored in a well laid out 'sequential binary file'. Been there, done that. Those types of files have a very well defined structure, and can be optimal for their intended use - eg. vertical seismic data, horizontal (time slice) data, horizon sliced data, and so forth.
Does that make them better ? I am sure an sql database could store the same information - but would it meet MY requirements for speed of access, smallest amount of disk space required, flexibility (with speed) of access ? From my experience, no way. The overhead of oracle alone is pretty high.
Complex databases have their uses, not denying that at all.
Simple storage schemes also have their uses.
Is oracle better than Access ? Well, access is sure a lot cheaper (and probably easier to use). *w*
Do the AI systems use sql databases to store their massive amounts of information ? I have no idea, but kinda suspect they've got an entirely different methodology, structure and access techniques.
Give me DMA transfers - I love 'em. *lol*
| 
