767
Implementing a SQL Data Warehouse
Exam number: 70-767
Exam title: Implementing a SQL Data Warehouse
Publish date:
GUID:
Language(s) this exam will be available in:
Audience (IT professionals, Developers, Information workers, etc.):
Technology:
Credit type (example: MCSA):
Exam provider (VUE, Certiport, or both):
Exam Design
This document shows changes to objectives and functional groupings. These
changes are effective as of March 2017.
Audience Profile
This exam is intended for extract, transform, and load (ETL) and data warehouse developers who create business
intelligence (BI) solutions. Their responsibilities include data cleansing as well as ETL and data warehouse
implementation.
Skills measured
Design, and implement, and maintain a data warehouse (35-40%)
Design and implement dimension tables
Design shared and conformed dimensions; determine support requirements for slowly
changing dimensions; determine attributes; design hierarchies; determine star or snowflake
schema requirements; determine the granularity of relationship by using fact tables;
determine auditing or lineage requirements; determine keys and key relationships for a data
warehouse; implement dimensions; implement data lineage of a dimension table
Design and implement fact tables
Identify measures; identify dimension table relationships; create composite keys; design a
data warehouse that supports many-to-many relationships; implement semi-additive
measures; implement non-additive measures
Design and implement indexes for a data warehouse workload
Design an indexing solution; select appropriate indexes; implement clustered, non-clustered,
filtered, and columnstore indexes
Design storage for a data warehouse
Design an appropriate storage solution, including hardware, disk, and file layout
Design and implement partitioned tables and views
Design a partition structure to support a data warehouse; implement sliding windows;
implement partition elimination; design a partition structure that supports the quick loading
and scale-out of data
Manage and maintain a SQL Data Warehouse
Manage queries by using labels; manage statistics; manage partition distribution; scale out
the data warehouse; grow, shrink, and pause the data warehouse
Extract, transform, and load data (40-45%)
Design and implement an extract, transform, and load (ETL) control flow by using a SQL Server
Integration Services (SSIS) package
Design and implement ETL control flow elements, including containers, tasks, and
precedence constraints; create variables and parameters; create checkpoints, sequence and
loop containers, and variables in SSIS; implement data profiling, parallelism, transactions,
logging, and security
Design and implement an ETL data flow by using an SSIS package
Implement slowly changing dimension, fuzzy grouping, fuzzy lookup, audit, blocking, non-
blocking, and term lookup transformations; map columns; determine the appropriate
transform object for a given task; determine appropriate scenarios for Transact-SQL joins vs.
SSIS lookup; design table loading by using bulk loading or standard loading; remove extra
rows or bad rows by using de-duplication
Implement an ETL solution that supports incremental data extraction
Design fact table patterns; enable Change Data Capture; create a SQL MERGE statement
Implement an ETL solution that supports incremental data loading
Design a control flow to load change data; load data by using Transact-SQL Change Data
Capture functions; load data by using Change Data Capture in SSIS
Debug SSIS packages
Fix performance, connectivity, execution, and failed logic issues by using the debugger;
enable logging for package execution; implement error handling for data types; implement
breakpoints; add data viewers; profile data with different tools; perform batch clean-up
Deploy and configure SSIS packages and projects
Create an SSIS catalog; deploy packages by using the deployment utility, SQL Server, and file
systems; run and customize packages by using DTUTIL
Integrate solutions with cloud data and big data
Integrate external data sources with SQL Server by using Polybase
Integrate with Hadoop; integrate with text files stored in the Azure Blob service; manage
external tables; access data in Hadoop databases with Transact-SQL; access data in the Azure
Blob service by using Transact-SQL; import data from Hadoop or blobs as regular SQL Server
tables; export data to Hadoop or the Azure Blob service
Extract, transform, and load data from SQL Data Warehouse by using Polybase
Integrate Azure SQL Data Warehouse with on-premises data warehouses; implement bi-
directional data synchronization between Azure and on-premises systems; load data into SQL
Data Warehouse from Polybase; design an incremental load strategy by using Polybase and
the Azure Blob service
Design and implement an Azure SQL Data Warehouse
Create a new Azure SQL Data Warehouse database by using the Azure portal; create an Azure
SQL Data Warehouse database by using Transact-SQL; select the appropriate method to load
data into Azure SQL Data Warehouse
Manage and maintain a SQL Data Warehouse
Manage queries by using labels; manage statistics; manage partition distribution; scale out
the data warehouse; grow, shrink, and pause the data warehouse
Build data quality solutions (15-20%)
Create a knowledge base
Create a Data Quality Services (DQS) knowledge base; determine appropriate use cases for a
DQS knowledge base; perform knowledge discovery; perform domain management
Maintain data quality by using DQS
Add matching knowledge to a knowledge base; prepare a DQS for data deduplication; create
a matching policy; clean data by using DQS knowledge; clean data by using the SSIS DQS task;
install DQS
Implement a Master Data Services (MDS) model
Install MDS; implement MDS; create models, entities, hierarchies, collections, and attributes;
define security roles; import and export data; create and edit a subscription; implement
entities, attributes, hierarchies, and business rules
Manage data by using MDS
Use MDS tools; use the Master Data Services Configuration Manager; create a Master Data
Manager database and web application; deploy a sample model using MDSModelDeploy.exe;
use the Master Data Services web application; use the Master Data Services Add-in for Excel;
create a Master Data Management hub; stage and load data; create subscription views