Hello,

I would like to share my bad and good experience in application's design and planning.

1)

Remember, if you work with WEB/EJB container's objects, never make assumptions.

This stuff always tends to behave against your common sense.

As example:

Those of you following the widley used OpenSource database Not Owned by Oracle - PostgresSQL anoounced its 9.0 ver release yesterday, 9.0 has promised to deliver quite a large set of enterprise class features, see announcment @: http://www.postgresql.org/about/news.1235

One more .Net driver  for connecting to MongoDB - NoRM - open-source library managed on GitHub.

Speciality of this project in linkage of .Net types and document model of MongoDB.

With NoRM very easy to manipulate with .Net type without necessity to copy them to/from documents for saving/resolving in/from DB like  MongoDB-CSharp Driver.

NoRM Home http://wiki.github.com/atheken/NoRM/

Sample

    var coll = (new Mongo(connString)).GetCollection<Product>();

    //create a new object to be added to the collection
    var obj = new Product();
    obj._id = ObjectId.NewObjectID();
    obj.Title = "Shoes";

    //save the object
    coll.Insert(obj);

    //find the object
    var obj2 = coll.FindOne(new { _id = obj._id}).First();

Large scale analysis solution must be able to provide very fast response time (<1sec) to some complex queries which are typically structured from a large fact table which is joined to many dimension tables and grouped-by them.

Using standard SQL databases such as MySQL InnoDB (as well as leading commercial vendors) will result in poor performance even when correct indexes are set up since in many case the database will be forced to perform unnecessary full table scans which will kill the performance if dimension tables are large.