My collaborator on some high-end computing work said that SS drives are wonderful, but when they fail, they fail completely and spectacularly. No warning, no recovery, no nuthin. Everything just goes black and that's it -- and chances are that happens on a day when you have critical work to do.
He uses a strategy that has been hinted at somewhere above -- use SSD's to hold operational stuff (Operating system, software/programs, temporary files), and high speed disks of traditional form to hold data (imagery, datasets, results of analyses). A weak link in the strategy is the transfer rate from/to the older spinning disks (He uses high RPM and a very fast bus/protocol of some kind -- used to be SCSI). Still, this is probably the best approach. And of course he incorporates redundant disk arrays and stuff like that as well.
I had my personal/home office machine set up similarly: SSD for OS and programs, HD for actual data. It works out nicely that large capacity HD's are not terribly expensive, while high capacity SSD's are crazy expensive.
