IBM Big Data Engineer Set 1

You can improve the performance by increasing the number of map tasks assigned to the load ✘ When loading large files the number of files that you load does not impact the performance of the LOAD HADOOP statement ✘ You can improve the performance by decreasing the number of map tasks that are assigned to the load and adjusting the heap size ✘ It is advantageous to run the LOAD HADOOP statement directly pointing to large files located in the host file system as opposed to copying the files to the DFSprior to load ✔

InfoSphere Streams can both read from and write data to HDFS ✘ The Streams Big Data toolkit operators that interface with HDFS uses Apache Flume to integrate with Hadoop ✘ Streams applications never need to be concerned with making the data schemas consistent with those on Hadoop ✘ Big SQL can be used to preprocess the data as it flows through InfoSphere Streams before the data lands in HDFS ✔

HBase is not supported as an import target ✘ Data imported using Sqoop is always written to a single Hive partition ✘ Sqoop can be used to retrieve rows newer than some previously imported set of rows ✘ Sqoop can only append new rows to a database table when exporting back to a database ✔

Spark SQL does not support HiveQL ✘ Spark SQL does not support ANSI SQL ✘ To use Spark with Hive, HiveQL queries have to rewritten in Scala ✘ Spark SQL allows relational queries expressed in SQL, HiveQL, or Scala ✔

Big SQL doesn't support stored procedures ✘ Big SQL can be deployed on a subset of data nodes in the BigInsights cluster ✘ Big SQL provides a SQL-on-Hadoop environment based on map reduce ✘ Only tables created or loaded via Big SQL can be accessed via Big SQL ✔

hive.exec.max.dynamic.partitions.pernode ✘ hive.exec.max.dynamic.partitions ✘ hive.exec.max.created.files ✘ All of the above ✔

join ✘ group ✘ expand ✘ transform ✔

Hue submits MapReduce jobs to Oozie ✘ No additional setup is required to secure your session cookies ✘ Hue applications require some code to be installed on the client ✘ The File Browser application allows you to perform keyword searches across your Hadoop data ✔

INT ✘ STRING ✘ STRUCT ✘ BOOLEAN ✔

Search and custom requirements ✘ Accuracy, search, and performance ✘ Adaptive weighting and standardization ✘ Empirical components, accuracy, and performance ✔