The Hitchhiker's Guide to the Galaxy revealed that 42 is what you get when you multiply 6 by 9 (which explains why things keep going wrong). Now that we all know that, we can discuss what to do in one of those situations when things go wrong. What do you do when something goes wrong with your parallel program? As the The Hitchhiker's Guide to the Galaxy tells us, Don't Panic! The IBM Parallel Environment for AIX provides a variety of ways to identify and correct problems that may arise when you're developing or executing your parallel program. This all depends on where in the process the problem occurred and what the symptoms are.
This chapter is probably more useful if you use it in conjunction with IBM Parallel Environment for AIX: Operation and Use, Vol. 1 and IBM Parallel Environment for AIX: Operation and Use, Vol. 2. So, you might want to go find them, and keep them on hand for reference.
Note: The sample programs in this chapter are available from the IBM RS/6000 World Wide Web site. See Accessing PE documentation online for more information.
Before continuing, let's stop and think about the basic process of creating a parallel program. Here are the steps, (which we have greatly abbreviated):
As with any process, problems can arise in any one of these steps, and knowing which tools to use to identify, analyze and correct the problem is the first step. The remainder of this chapter describes some of the common problems you might run into, and what to do when they occur. The sections in this chapter are labeled according to the symptom you might be experiencing.