It's probably worth your while to take a loop at Hadoop them. When you can fundamentally map your problem into many small problems and then reduce the results

It's probably worth your while to take a loop at Hadoop them. When you can fundamentally map your problem into many small problems and then reduce the results (true for most sequence alignment, etc problems), then Hadoop is very viable. Not exactly common fare these days, but getting more popular and perhaps increasingly so. Plus you don't really have to learn new languages, just a distributed computing framework.