Hi Randy,

When a query is made against a MSSQL table, whether from the command line or by SPT, the MSSQL usually must parse the request. Also, MSSQL would run an execution plan to formulate the best approach/indexes to use in returning any results. Would both of these steps be necessary in making a call to a SP? If not, what is the cost/elape time expense difference between the two methods. Would it be significant. How does MSSQL compare to Oracle and DB2 in terms of cost/elape times to perform a parse of the SQL and to formulate an execution plan. This would not apply in all databases. For example, it is my understanding that Postgres does not parse anything. Also, MySQL does not have stored procedures. MySQL, to a limited extend, will perform an execution plan, but often needs help from the program/programmer in this area. The common denominator between all the database back-end servers seems to be compilance with SQL 92 and SQL 99 standards, so theortically, I could use any front-end application to connect to any back-end database server. As long as the SQL submitted to the back-end server adhered to SQL 92 or SQL 99 standards, the back-end sever really shouldn't matter. Of course each database back-end server has extended its functionality beyond SQL 92 and SQL 99, so these differences would need to be addressed by any front-end app. For example, some back-end databases, like MySQL, have a built-in auto-increment, while others, like Postgres, use a sequencing function to perform the auto-increment to obtain a unique primary keys. The syntax in addressing these two methods would be different.