A score of 26 of the 32 is normally considered a passing score. Use this information to assess your readiness to take the Teradata Certification Exams. When you are satisfied with your answers, click on the SUBMIT button at the end of the test. You will receive immediate notification of your score. 1. In creating the ELDM, which three data demographics are added to the model? (Choose three.) typical rows per value maximum rows per value sample data distinct values 2. The requirement is to load a small amount of control data stored in an MSAccess database into Teradata. Which application tool should you use to load the data? WinCLI PP2 BTEQ OLE/DB 3. What does the Teradata Index Wizard do? (Choose three.) creates a view of your data demographics allows importing of workload from Teradata Manager allows for a visual explanation of the query plan allows you to schedule index creation allows for What-If analysis 4. Why are row-hash level locks preferable in tactical queries? reduced occurrence of deadlocks blocked queries spend less time in the lock queue promote higher query concurrency only affected rows are locked 5. When performing a row insert on a SET table, when is performance overhead added to the operation? (Choose four.) when soft referential integrity is defined on the table when a NUSI is defined on the table when a NUPI is defined on the table when batch referential integrity is defined on the table when a PPI is defined on the table 6. A query that performs identical steps across multiple parallel units is an example of what kind of parallelism? multi-session multi-step multi-statement multi-AMP multi-threading 7. Which two kinds of applications should use a macro? (Choose two.) OLTP-type of application one that contains repeated SQL statements with only minor differences one that uses few AMPs one that does a primary index retrieve 8. Which two data types hash the value 100 identically? (Choose two.) INTEGER FLOAT DECIMAL(9,0) DECIMAL(12,1) 9. What are three advantages of defining a Partitioned Primary Index (PPI)? (Choose three.) enables more even distribution of rows to the AMPs provides fast delete of whole partitions provides fewer rows per hash value within a partition provides access via a constraint on the partitioning columns 10. What determines the time of day by which data loads must be completed? nightly batch window source data availability impact on system resources business requirements 11. Which two metrics are used when planning database capacity for new applications? (Choose two.) network bandwidth data model data volumes the number and type of queries 12. Which is a characteristic of quality assurance (QA) testing? It must be performed on the production system. The data used need not be representative of production data. It is the on-going evaluation once the application is in production. It is a staging area between test and production. 13. What are two characteristics of a trigger? (Choose two.) CREATE TABLE is a valid statement as a triggered event. Tables referenced by triggers cannot be dropped unless the triggers have been removed. FastLoad or MultiLoad cannot be used on tables that have triggers. Triggers can be used to enforce RI. 14. What are three characteristics of views? (Choose three.) They can be defined with a locking modifier. They cannot contain an ORDER BY clause. They can be used to insert or update data in the underlying table. They require the user to have privileges on the underlying tables. They cannot reference Global Temporary tables and Volatile tables. 15. You have a requirement to use ANSI mode. You are implementing a multiple-request transaction. A failure (abort) occurs on the last request that contains a DDL statement. Which statement is true? Only the request that fails is rolled back. The code should be modified to include BT/ET. Locks for only the failed request are released. All requests in the transaction are rolled back. 16. Which three statements about the outer join are true? (Choose three.) SELECT C10, C20, C30 FROM T1 LEFT JOIN T2 ON T1.C2 = T2.C2 AND T2.C8 = 'xyz' WHERE T1.C9 = 'abc'; Table T2 is called the inner table. The AND condition is residual to the inner join. The WHERE condition is logically applied first. The ON clause defines the inner result 17. An input parameter is evaluated to determine which of three tables is to be scanned. The table is scanned, a result is computed and returned to the user. Which type of application implementation should be used to achieve this objective? view macro stored procedure trigger 18. What are four limitations of the FastLoad utility? (Choose four.) It can use only a single input data stream. The target table must be empty. Error table 1 will include the relative input record number. It allows duplicate row processing into the database. Only a single table can be loaded. Secondary indexes or referential integrity cannot be present. 19. Which three statements about FastExport are true? (Choose three.) It supports the SELECT and DELETE DML statements. It supports DDL statements. It supports the same export data modes as BTEQ. It supports multiple sessions. It supports INMODs and OUTMODs. 20. Which statement about Standard Referential Integrity is true? Circular referential constraints are not allowed. Referential constraints do not allow null value in a column. Referential constraints must be established on empty tables or at table create time. Referential constraints must be created at either the column or the table level. 21. Which three LOAD utilities can be used to add the value of two input fields into one table field? (Choose three.) Multiload TPUMP Fastload using EXIT routine Fastload using INMOD 22. Which statement about referential integrity is true? All forms of referential integrity require the creation of an error table. Neither the WITH CHECK OPTION nor the WITH NO CHECK OPTION enforce integrity when inserting rows into the referencing table. All forms of referential integrity require a unique index on the referenced column list. The Optimizer can only utilize join elimination for referential integrity when no reference is made to WITH CHECK OPTION or WITH NO CHECK OPTION. 23. You read an EXPLAIN and notice that the Optimizer is duplicating a table on all AMPs. Which three are possible join plans for this step? (Choose three.) nested hash merge product row-id 24. Which distribution strategy is chosen by the Optimizer when the join calls for a cross join? choose the smaller spool file, duplicate it on all AMPs, and locally copy the other table into a spool file to be prepared for the join join by accessing the table rows without redistribution send all rows of the smaller table to a single AMP for processing and locally copy the larger table into a spool file on each AMP choose the smaller spool file and duplicate it on all AMPs leaving the other table in place redistribute one or both by the column that is the join constraint 25. What are three possible join strategies available to the Optimizer? (Choose three.) Exclusion Product Join Hash Join Exclusion Nested Join Inclusion Merge Join 26. Which type of query conditionally executes steps based on existence of nulls? exclusion merge join inner joins on nullable columns nullable NOT IN subquery multi-column aggregations 27. What does a pseudo-table lock avoid? deadlocks AMP lock row level lock hash collisions 28. Which three statements are true for sampled statistics collection? (Choose three.) Sampled and complete statistics are both stored for the same index or column set. Sampled statistics are more appropriate for non-indexed columns than indexed columns. Sampled statistics reduce resource consumption from the collection process. Sampled statistics are more accurate for data that is uniformly distributed. Sampled statistics may negatively affect the EXPLAIN plan. 29. Which statement applies to a value ordered index? It can be unique. It can be any integer data type. It cannot be updated via MultiLoad. It can be any data type of four bytes or less. It can be a primary index. 30. Which statement is true for a full covering join index? The full covering index is used if all columns selected are in the index. The full covering index is used if most of the columns selected are in the index. The full covering index is used if the primary index of both tables are in the covering index. The covering join index is used if the join index is an aggregate. 31. What are three possible benefits of a Partitioned Primary Index (PPI)? (Choose three.) avoids full-table scans avoids secondary indexing improves full-table joins of tables with same primary index improves deletion of data 32. Which three let you monitor or record SQL requests sent to the database? (Choose three.) Database Query Log (DBQL) Teradata Administrator Teradata Manager Teradata Dynamic Query Manager (TDQM)