Gerry Snyder wrote: > steffen scheer wrote: >> Hey all >> >> I want to use sqlite as a storage engine for DNA sequence data. It is >> part of a workflow engine for DNA processing. I was wondering whether i >> chose the right db design. >> >> The db holds information about DNA sequences. Arranged in three tables. >> For every dna sequence one entry in all 3 tables will be made (1:1 >> Relation). Data will be inserted once, never deleted, but some >> attributes get updated. >> > To me, KISS (keep it simple, stupid) rules until its performance has > been shown to be unacceptable. > > Surely you can create a dummy db file with nonsense data of typical size > all in a single table and see whether it meets your needs. > > Keeping 2 or 3 tables in synch is not a difficult task, but it is a > potential source for error.
You are right. I merged seq_desc_static and seq_desc_var. This was my initial design. In the majority of cases the first design is the best. KISS rules Steffen _______________________________________________ sqlite-users mailing list [email protected] http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
