|
You’ve tested it to destruction but you still can’t answer the question "HOW WILL IT BEHAVE IN PRODUCTION ?"
SIMULOAD from Oriole can tell you. You know how it is, developers write code which, give or take a few bugs, works. The snag is that they usually test it on small tables, creating the 'New York' and 'Tokyo' databases on the same machine and checking how they communicate. Not very realistic is it? If you’re lucky they run the product against a copy, rather than a subset of production data, only to find that response times are awful. But whatever they do, they can never properly simulate a large number of concurrent users. As a result, the product looks OK, the demos look OK, functionalities are tested, the product goes into production and after a week you're in total crisis and a succession of tempestuous meetings because users have to wait 20 seconds for the screen to refresh. SIMULOAD allows you to simulate both OLTP and batch programs – and the interactions between the two, with any number of concurrent users. SIMULOAD allows you confidently to predict whether your current hardware configuration and programs will be able to serve a larger number of Oracle users and/or processes. SIMULOAD will tell you how increased load will affect response times of end-users (and top management?). SIMULOAD is simple to use and inexpensive Anyone who knows SQL, can use SIMULOAD. It takes as input, a very simple parameter file, shell scripts and/or SQL queries or PL/SQL programs. You can even link your Pro*C programs with SIMULOADPRO if you want very precise figures. SIMULOAD is a very powerful tool which will pay for itself over and over again and has been realistically priced to achieve maximum market penetration. SIMULOAD comes in three classes at $2,500, $4,000 and $5,000 so, as you can see, if your organization can’t afford it, then it can’t really need it. |
Why leave it to trial and error?