According to EJB spec, if table has one million rows in the table, then it will create one million entity bean instances for each row in the table. Apart from object pooling, if different clients require the one million different data at the same time, then it will create entity bean instances for each request. Creating one million objects, is a big performance issue. Could you please any one give me the exact picture of it.
It is already using object pooling to control it, one instance is not used then it will passivity and activate later when it needs. My question is, if one million clients are requesting different data(different rows in the table) concurrently at the same time, it has to create one million entity bean instances. It is a very big performance issue.