Prepare your SMS testing environment and create the test case library.

1. Perform data classification to determine the data types.

   While this step is optional, if you do not perform data classification for the data types, you must still identify the data subtypes that you want NaviQuest to test.

   _______________________________________________________________

2. Assign the subtype prefix.

   If you performed data classification (step 1), you must assign a 1-to-4 character member name prefix to each data classification data subtype. Even if you do not use data classification, you still must identify the different types of tests you want done and assign them a subtype prefix, because each test runs against a different data subtype and produces a different expected result.

   Requirement: For the base line test set, you must use the subtype prefix NEVR.

   _______________________________________________________________

3. Determine the amount of testing to be done.

   You can choose to conduct tests of all data sets in a given data subtype or you can conduct tests with a selected percentage of the data sets. If you use all the data sets, the testing is more extensive, with less chance for error. However, you can use a sample of the data to minimize the resources used while testing.

   _______________________________________________________________

4. Determine when to place test cases into the testbed library.

   The testbed library stores test cases for later use in bulk or regression testing.

   You can either initially place test cases for all data subtypes into the testbed library or wait to add each subtype test set to the testbed library at the start of each phase. Fully loading the testbed library with all test cases prior to any testing results in more thorough testing. Alternatively, adding the test cases at the start of each phase reduces the total time taken for test runs.

   _______________________________________________________________

5. Determine how the data for the test cases is to be created.

   You must also decide how you want to collect the data set test cases. For each data subtype, decide if you want ISMF lists, DCOLLECT, VMA, or SMF to be the source of the test cases. See Methods for collecting test data for an explanation of the data collection methods.

   The percentage of test data and method of collecting test case data can be tailored for each data subtype. This is independent of your choice of storing all test cases as you initially build the testbed library or at the start of each phase’s testing. For example, you may decide to test all production data sets for production data, but to use only a ten percent sample of your test data sets.

   _______________________________________________________________

6. Determine the order of testing.
   Because the order of testing is normally the same as the order of data conversion to SMS management, you must determine the order in which you want to test data subtypes with NaviQuest. Only a single data subtype can be initially tested at one time.

   Tip: Other data subtypes that have already been tested are regression tested at this time. NaviQuest testing is usually done against all subtypes of one data type before starting the next.

   _______________________________________________________________

7. Allocate the testbed library.

   You can either create a testbed library or let NaviQuest create one for you. The following instructions will guide you in how to create one.

   To create a testbed library, create a PDS with the following attributes:

   • LRECL = 80
   • BLKSIZE = 0

   Each test case is a member of the PDS. All test cases associated with a single data type have the same subtype prefix and have the identical expected results.

   The amount of space allocated to the library depends on how many data sets the installation has and how many test cases you want to create. For example, on a 3380 device, 1000 test cases take approximately 8 cylinders and 52 directory blocks.