最新的Microsoft Implementing Analytics Solutions Using Microsoft Fabric (DP-600 Korean Version) - DP-600 Korean免費考試真題

說明：（僅 Fast2test 成員可見）

Explanation:

說明：（僅 Fast2test 成員可見）

Explanation:

Step 1 - Read the Parquet file into a DataFrame
df = spark.read.parquet( " abfss://[email protected]/files/sales.parquet " ) This correctly loads the Parquet data into Spark.
Step 2 - Write into the Lakehouse as a managed table
If we want the result to be registered as a Lakehouse table and automatically appear in Lakehouse Explorer, we must:
Write the data in delta format (because Fabric Lakehouse tables are Delta tables).
Save the table under the tables folder, not files.
So the correct code is:
df.write.mode( " overwrite " ).format( " delta " ).saveAsTable( " tables/sales " ) Final Answer:
Format: delta
SaveAsTable Path: tables/sales
References:
Lakehouse tables in Microsoft Fabric
Save DataFrame as Delta Table in Spark
# Answer Selection:
First dropdown # delta
Second dropdown # tables/sales

Explanation:
For the HighestSellingPrice, you should use the GREATEST function to find the highest value from the given price columns. However, T-SQL does not have a GREATEST function as found in some other SQL dialects, so you would typically use a CASE statement or an IIF statement with nested MAX functions. Since neither of those are provided in the options, you should select MAX as a placeholder to indicate the function that would be used to find the highest value if combining multiple MAX functions or a similar logic was available.
For the TradePrice, you should use the COALESCE function, which returns the first non-null value in a list.
The COALESCE function is the correct choice as it will return AgentPrice if it ' s not null; if AgentPrice is null, it will check WholesalePrice, and if that is also null, it will return ListPrice.
The complete code with the correct SQL functions would look like this:
SELECT ProductID,
MAX(ListPrice, WholesalePrice, AgentPrice) AS HighestSellingPrice, -- MAX is used as a placeholder COALESCE(AgentPrice, WholesalePrice, ListPrice) AS TradePrice FROM Sales.Products Select MAX for HighestSellingPrice and COALESCE for TradePrice in the answer area.

說明：（僅 Fast2test 成員可見）

說明：（僅 Fast2test 成員可見）

Explanation:

GRANT ALTER ON SCHEMA::schemaA TO User1;
The ALTER permission allows a user to modify the schema of an object, and granting ALTER on a schema will allow the user to perform operations like TRUNCATE TABLE on any object within that schema. It is the correct permission to grant to User1 for truncating tables in schemaA.
References =
GRANT Schema Permissions
Permissions That Can Be Granted on a Schema

說明：（僅 Fast2test 成員可見）

說明：（僅 Fast2test 成員可見）

Explanation:

Comprehensive Detailed Explanation
We have a KQL table (TaxiData) with columns:
VendorID
tpep_pickup_datetime (timestamp)
payment_type
total_amount
The requirement:
Create a new column FirstPickupDateTime
It should contain the first pickup timestamp per hour
Partitioning should be done by payment_type
Step 1: Which windowing function?
row_cumsum # running cumulative sum (not needed here).
row_rank_dense # assigns ranks without gaps, but does not guarantee minimum value only.
row_rank_min # gives the first/minimum value in each window partition. # Correct.
row_window_session # sessionization of events, not required.
So, the correct function is row_rank_min.
Step 2: Which comparison operator?
We need to select the row where the rank = 1 (the first per partition).
So the correct operator is == (equals).
Step 3: Partitioning
The KQL query should partition by:
bin(tpep_pickup_datetime, 1h) # buckets data into 1-hour windows
payment_type # partitions further by payment type
Completed KQL Query
TaxiData
| sort by tpep_pickup_datetime asc, payment_type asc
| extend FirstPickupDateTime = row_rank_min(tpep_pickup_datetime, 1h, 0m, payment_type)
| where FirstPickupDateTime == 1
This assigns a rank within each 1-hour, per-payment-type window, then keeps the first pickup timestamp.
Why This Works
row_rank_min # ensures we capture the first occurrence in each hour.
== # filters only the first row per partition.
bin(..., 1h) ensures grouping is by hour.
References
Kusto row_rank_min() function
KQL window functions

Explanation:
The results will form a hierarchy of folders for each partition key. - Yes The resulting file partitions can be read in parallel across multiple nodes. - Yes The resulting file partitions will use file compression. - No Partitioning data by columns such as year, month, and day, as shown in the DataFrame write operation, organizes the output into a directory hierarchy that reflects the partitioning structure. This organization can improve the performance of read operations, as queries that filter by the partitioned columns can scan only the relevant directories. Moreover, partitioning facilitates parallelism because each partition can be processed independently across different nodes in a distributed system like Spark. However, the code snippet provided does not explicitly specify that file compression should be used, so we cannot assume that the output will be compressed without additional context.
References =
DataFrame write partitionBy
Apache Spark optimization with partitioning