Expect the IO team to be defensive and possibly very aggressive in response to this news. After all, when millions of dollars are spent on an IO subsystem and the average write takes 50 ms, people get very nervous. Keep stressing that Oracle's sequential IO write requirements have simply exceeded the IO subsystem's capacity, and that people are working to reduce the IO requirements by focusing on Oracle and on the application. You can also show them the IO requirements Oracle is placing on the IO subsystem. Reports similar to the ones shown n Figures 8-10, 8-15, and 8-16 can be very helpful.

Oracle-focused solutions center on reducing redo generation and ensuring the log writer background process is writing efficiently. We want to make sure that when the log writer background process takes the time to issue a write request, it pushes down a large batch of redo. It is much more efficient to perform a single 1MB IO write than to repeatedly perform 512 writes of 2,048 bytes.

As I mentioned in the previous section about application-focused solutions, seriously consider finding creative ways to implement Oracle's global temporary tables. If the redo is associated with a restartable batch process, there is a good possibility of success. Using global temporary tables can cut redo generation by 50%, substantially reducing Oracle's overall IO requirements. This alone could solve the log file parallel write problem.