From: esquel@sommarskog.se   
      
   M.G. (michael@gurfinkel.us) writes:   
   > We are designing a table with high insert / delete activity. The table   
   > maintains sequence of actions per specific experiments. These are the   
   > attributes:   
      
   Important question: is the activity concurrent or not? That is, can data for   
   multiple experiments be written simultaneously?   
      
   > Expected number of records - around 10 million, expected number of   
   > inserts (pure additions) around 5000 a day, number of change   
   > (delete/insert) around 1000   
      
   OK, so that is not really high activity. :-)   
      
   And then I would assume that there is no concurrent activity.   
      
   I would say that you scrap the ACTIVITY_SEQUENCE column, as it seems   
   to serve no purpose. Make (EXP_ID, ACT_SEQ, ACT_ID) a primary key, because   
   a having a primary key based on data is a good thing. You may never   
   look at indivitual data, but you don't want duplicates. And make   
   this PK clustered, that seems to fit your use of the table very well.   
      
      
      
      
   --   
   Erland Sommarskog, Stockholm, esquel@sommarskog.se   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)