A00-221: SAS Big Data Programming and Loading Using SAS 9

Exam ID: A00-221

Exam Name: SAS Big Data Programming and Loading Using SAS 9

Successful candidates should have hands-on experience with a variety of SAS® and open source data preparation tools, including:

Try Online Exam »

  • SAS Base and DS2 programming.
  • Hadoop, Apache Pig, Apache Hive.
  • SAS In-Memory Statistics.

SAS A00-221 Exam Summary:

Exam Name SAS Big Data Programming and Loading
Exam Code   A00-221
Exam Duration   105 minutes
Exam Questions   60 to 65 Multiple choice questions
Passing Score   68%
Exam Price   $180 (USD)
Books  1. SAS Academy for Data Science: Big Data
2. DS2 Programming Essentials with Hadoop
3. Introduction to SAS and Hadoop
4. Getting Started with SAS In-Memory Statistics
Sample Questions   SAS Big Data Professional Certification Sample Question
Practice Exam   SAS Big Data Professional Certification Practice Exam

SAS A00-221 Exam Topics:

Objective   Details
SAS and Hadoop – 30%
Describe the baseline requirements for interacting with Hadoop – SAS_HADOOP_JAR_PATH and SAS_HADOOP_CONFIG_PATH environment variables
– JAR file requirements
– Hadoop XML configuration file contents
– JAR files and XML files must be accessible to the SAS Server
– Understand the precedence of settings for Hadoop XML configuration files
– Identify the components of a SAS and Hadoop solution
– List the communication paths between the components of a SAS and Hadoop solution
Use the HADOOP procedure and the Hadoop FILENAME statement to interact with Hadoop from a SAS session – Know which HDFS commands are available through the Hadoop procedure
– Submit HDFS file system commands (DELETE, MKDIR, RENAME, CHMOD, LS, CAT)
– Copy files between SAS and Hadoop via COPYFROMLOCAL and COPYTOLOCAL statement
– Submit MapReduce programs with the MAPREDUCE statement
– Understand best practice considerations when using the FILENAME statement
– use the FILENAME statement to read data from and write data to the Hadoop file system in a SAS DATA step
– Execute Pig code with the PIG statement in the HADOOP procedure
Query and manage Hive tables stored in Hadoop using explicit SQL pass-through – Manage connections to Hive with the CONNECT/DISCONNECT statements (schema, server, username, password, etc)
– Access Hive metadata via SHOW and DESCRIBE statements
– Select data from tables with HiveQL (select, from, where clauses)
– Join tables with HiveQL
– Use both HiveQL and SAS SQL features (ORDER BY, functions, labels) in the same SQL procedure SELECT statement
– Create SAS data sets and views from Hive results
– String dates vs. SAS dates
– Using the CAST function to control data type in explicit queries (32k string lengths)
– Create Hive table definitions
– Load data into Hive table defitions from local data
– Load data into Hive table definitions from HDFS data
– Control length of character variables created in Hive tables
– Work with Hive string types in SAS
– Control Hive table properties with TBLPROPERTIES statement (SASFMT or with data set option DBSASTYPE= option)
– Compare managed and external Hive tables
– Compare different Hive file types (textfile, sequencefile). Use of SERDEs
– Use data set options to define specific HDFS file types
Work with Hadoop files using the SAS/ACCESS LIBNAME statement – Write a LIBNAME statement to access Hive tables
– Access Hive metadata using LIBNAME statement and the CONTENTS procedure
– Understand that the SAS/ACCESS engine writes database-specific SQL code when using implicit pass-through
– Maximize the use of HiveQL by optimzing implicit pass-through (summarization, subsetting, joins)
– Use system options to determine where processing occurs (SASTRACE, NOSTSUFFIX, SASTRACELOC)
– Evaluate SAS logs to determine the amount of implicit pass-through performed by a SAS program
– Identify SAS data set options that can be implicitly passed to Hive
– Identify SAS functions that can be passed to Hive
– Convert date formats with the SASDATEFMT= data set option
– Embed LIBNAME statements in SQL View defintions
– Identify best practices when combining tables to maximize Hive usage
– Methods to combine/join tables
– Copy data sets to Hive using the COPY procedure
– Identify advantages of using SAS/ACCESS LIBNAME method
– Identify disadvantages of using SAS/ACCESS LIBNAME method
– Maximize performance when using the LIBNAME statement
– Efficient methods for BY GROUP processing with in-database procedures
– Managing data types for computed columns.
– Partition and cluster Hive tables
– Create Hive external tables
SAS DS2 Programming – 30%
Write DS2 programs – Utilize run group processing
– Use DATA, ENDDATA, and RUN statements properly
– Use system methods, INIT(), RUN(), TERM()
– Build user defined methods
– Pass arguments to user defined methods
– Explain the use of the INIT(), RUN(), TERM() system methods
– Use the OVERWRITE option
– Understand how DS2 handles reserved keywords
– Recognize components of traditional DATA Step programming that are or are not supported in DS2
Read data using DS2 – read data with a SET statement
– write FedSQL code within SET statements to read data
– Use FedSQL SELECT statements to extract specific variables from input data sets
– Use FedSQL Join statements to merge data from multiple input data sets
– Use FedSQL WHERE statements to extract specific observations from input data sets
– Use the MERGE statement to join data.
– Subset data using subsetting IF statements
– Read table data with a BY statement, without pre-sorting the data
– Use a FedSQL query with an ORDER BY clause to provide sorted data to the SET statement for BY group processing
Work with variables, arrays, and ANSI SQL data types – Define and use local and global variables (understand scope, what goes into PDV, output data sets)
– Declare variables with the DCL statement
– Use fractional, integer, character and Date & Time ANSI SQL data types
– Use CHAR, NCHAR, VARCHAR, NVARCHAR character data types
– Use DECIMAL, DOUBLE, FLOAT, REAL fractional numeric data types
– Use BIGINT, INTEGER, SMALLINT, TINYINT interger numeric data types
– Use BINARY, VARBINARY binary data types
– Use DATE, TIME, TIMESTAMP date and time data types
– Identify coercible and non-coercible data types
– Understand autoconversion of DS2 data types when DS2 variables are output to SAS data sets
– Select variables with KEEP and DROP statements and KEEP= and DROP= options
– Understand how SAS will perform automatic type conversions
– DS2SCOND option statement
– Use ANSI quoting standards in variable assignment statements
– Use macro variables within ANSI quoted variable assignment statements (%TSLIT macro)
– set variable attributes (Length, format, informat) within variable declaration statements
– Use the VARARRAY statement to declare arrays
– Use the DCL staement to declare temporary arrays
– Assign values to array variables
– Understand the difference between SAS MISSING and ANSI NULL data values
– Invoke ANSI data processing mode for NULL values with the ANSIMODE option
Use expressions and functions in DS2 programs – Use the DS2 IF expression in place of IF/THEN conditional statements
– Use the DS2 LIKE expression to compare character values to specific patterns
– Convert SAS datetime variables to DS2 ANSI TIMESTAMP variables with the TO_TIMESTAMP function
– Convert SAS date variables to DS2 ANSI DATE variables with the TO_DATE function
– Convert SAS time variables to DS2 ANSI TIME variables with the TO_TIME function
– Convert DS2 ANSI DATE, TIME, and TIMESTAMP variables to SAS date, time and datetime variables with the TO_DOUBLE function
– Increment date and time values with the INTDT and INTTS functions
– Execute FedSQL statements with the SQLEXEC function
Work with Methods, Packages, and Threads – Create methods that modify parameters at the site by using IN_OUT variables
– Create methods that return a value using a RETURN statement
– Overload methods by creating methods with multiple signatures
– Create user defined packages with the PACKAGE statement
– Understand the capabilities of predefined DS2 packages (such as FCMP, SQLSTMT, HASH, JSON)
– Intantiate DS2 packages with the DECLARE statement
– Use threading to alleviate CPU bound processes
– Create threads using the THREAD statement
– Declare instances of threads in a DS2 program
– Call threads using a SET FROM statement
– Specify the number of threads using a THREADS= option
– How to run threads inside parallel databases
– DS2ACCEL = YES option
– Requirements to execute DS2 code in-database
Hadoop Programming – 15%
Describe the Hadoop architecture – Identify Hadoop elements such as Name Nodes, Data Nodes, Job Trackers, Task Trackers, YARN
– Explain Hadoop concepts such as distributed storage & processing, splits, replication, MapReduce
– Describe components of the Hadoop ecosystem (Hive, Pig, Sqoop)
– Describe attributes of big data
– Identify use cases for Hadoop
Manipulate and load data files using command line tools – Use Linux shell commands (ls -ltr/pwd/mkdir/cd)
– Load and manipulate data into Hadoop using Linux commands (hdfs dfs – mkdir/put/copyFromLocal/ls/cat)
– Use Sqoop to move data from a RDBMS into Hadoop
Write Hive programs to create, join, and query data tables – Create databases and tables in Hive
– Understand the difference between external and internal tables
– Work with Hive variable types
– Load data into Hive table definitions with LOAD and INSERT statements
– Recognize challenges when importing data (embedded delimiter characters, header values)
– Limit values returned by a Hive query with the SELECT…..LIMIT keyword
– Sort Hive query results with the SELECT…ORDER BY keyword
– Group Hive query results with the GROUP BY keyword
– Choose which values to select from a data table using the SELECT WHERE keyword
– Retrieve unique values with the SELECT DISTINCT
– Join tables (Inner, Outer, Left, Right)
– Use functions in Hive queries (sum, count, avg, max, min, round, floor, ceil, rand, concat, substr, upper, ucase, lower, lcase, trim)
– Use relational and arithmetic operators in Hive queries
Write Pig programs to perform ETL tasks and to analyze large data sets – Identify Pig data types
– Use name and positional references in Pig programs
– Identify valid identifiers (start with letter, then letters, digits, underscores)
– Use Arithmetic, String, and Boolean Expressions
– Use the CAST operator to change variable types
– Increase parallel processing with the PARALLEL keyword
– Combine data from multiple tables with INNER, LEFT, RIGHT, and OUTER JOIN keywords
– Combine data using special join types: REPLICATED, SKEWED, MERGE
– Use parameters in a Pig program
– Use Diagnostic operators: DESCRIBE, EXPLAIN, DUMP, ILLUSTRATE
– Use functions in Pig programs
– DATE Functions: CurrentTime, DaysBetween, HoursBEtween, GetDay, GetHour (etc.), AddDuration, ToUnixTime, ToDate, ToMilliSeconds, ToString
– Tuple, Bag, Map functions: TOTuPLE, TOBAG, TOMAP, TOP
– Register and use User Defined Functions 
Data manipulation with the IMSTAT procedure – 25%
Execute IMSTAT procedures – Define a SASIOLA library to access in-memory data in a LASR Analytic Server
– Describe the key functionality of the IMSTAT procedure
– Perform one-dimensional numerical exploration with IMSTAT procedure statements SUMMARY and FREQUENCY
– Perform two-dimensional numerical exploration using the CROSSTAB or GROUPBY=option 
Perform actions required to produce graphs with PROC IMSTAT – Use PROC IMSTAT statements and options that calculate summary statistics for graphing
– Transfer the summary statistics tables to the SAS server 
Manipulate In-Memory Data – Define WHERE clauses to explore subsets of an in-memory table
– Create permanent columns using the COMPUTE statement
– Create temporary columns using temporary expressions of computed columns
– Work with SAS formats in the IMSTAT procedures
– Use the Fetch statement to retrieve data from an in-memory table
– Join in-memory tables 
Use High-Performance procedures with the SAS LASR Analytic Server – Compare the SAS High-Performance procedures and SAS IN-Memory Statistics
– Use the HPIMPUTE procedure to add imputed columns to an in-memory table