Microsoft DP-750 Exam Dumps

CORRECT ANSWE R: A - INSERT INTO sales.schema1.table2 SELECT * FROM sales.schema1.table1;

According to Microsoft Learn on Delta Lake DML operations, INSERT INTO appends all rows from the source table to the target table without affecting existing data. This directly satisfies both requirements: 'preserve any existing data in table2' and 'minimize processing effort.' Option B (CREATE TABLE AS SELECT) would create a new table or fail if table2 already exists --- it does not preserve existing data. Option C (INSERT OVERWRITE) replaces all existing data in table2 with the rows from table1, violating the 'preserve existing data' requirement. Option D (CREATE OR REPLACE TABLE AS SELECT) drops and recreates table2 entirely, deleting all existing historical data. INSERT INTO is the simplest, most direct command for appending data from one Delta table to another while preserving all existing rows.

CORRECT ANSWE R: A - merge.

According to Microsoft Learn on Git operations in Databricks Repos, the Git merge command combines the commit history of two branches and preserves all individual commits from both branches in the repository history. This satisfies the requirement to 'preserve the commit history in the repository.' A merge creates a merge commit that joins the two branch histories, keeping all individual commits from Branch1 visible in the log. Option B (pull) fetches remote changes and merges them but is used to sync with a remote, not to incorporate a feature branch into main. Option C (rebase) rewrites the commit history by replaying Branch1 commits on top of main, which does NOT preserve the original commit history (it creates new commit hashes). Option D (push) sends local commits to the remote but does not incorporate branch changes.

CORRECT ANSWE R: D - Apply the main branch changes to Branch1 and resolve the conflicts.

According to Microsoft Learn on Databricks Git integration and standard Git workflows, when a merge fails due to conflicts, the recommended approach is to pull/rebase the main branch changes into the feature branch (Branch1), resolve the conflicts locally within Branch1, and then merge the conflict-free branch into main. This ensures Notebook1 includes all changes from both branches. Option A (clone Branch1 as new repository) creates a disconnected copy and does not resolve the merge conflict. Option B (apply changes directly to main) bypasses the branch protection model and risks overwriting the other user's changes. Option C (clone main as new repository) similarly bypasses proper conflict resolution. The correct Git practice is always to resolve conflicts in the feature branch before merging into main.

CORRECT ANSWE R: D - Add a NOT NULL constraint to transaction_id and a CHECK constraint to amount.

According to Microsoft Learn on Delta Lake table constraints, Delta tables support two types of data quality constraints: NOT NULL constraints and CHECK constraints. A NOT NULL constraint ensures that the specified column cannot contain null values --- applying this to transaction_id satisfies 'transaction_id must never be null.' A CHECK constraint enforces a boolean expression on a column --- applying CHECK (amount > 0) ensures that only positive amounts are written. When a row violates either constraint, Delta Lake rejects the write operation, satisfying 'invalid records must be rejected when data is written to the Sales table.' These are table-level constraints enforced by the Delta engine on every write. Option A (SELECT with WHERE) is a read-time filter, not a write-time constraint. Options B and C (RLS and views) are read-time access controls, not write constraints.

CORRECT ANSWE R: C - Grant the analysts the SELECT permission for the Transactions table and apply column masks to email_address and credit_card_number.

According to Microsoft Learn on Unity Catalog column masking, column masks are the correct feature for partially obscuring sensitive column values while still allowing queries to return results. A column mask is a SQL function applied to a column that transforms the output based on the user's identity or group membership. In this scenario, the mask for email_address would return only the part after '@', and the mask for credit_card_number would return only the last 4 digits. Option A is incorrect because row-level filters restrict which rows are visible, not which column values are partially visible. Option B is incorrect because granting SELECT only on non-sensitive columns would prevent analysts from seeing any email or card data at all, which doesn't meet the 'can see partial values' requirement. Option D is incorrect because column-level encryption would encrypt the data and require decryption keys, adding significant development complexity beyond column masking.