In the Introduction to Administering Hadoop Clusters training course, you'll learn:
Hadoop Concepts
- Operate on large data sets
- Parallelize to improve performance
- Use large block sizes
- Distribute and replicate data
- Assign code to data
- Compensate for Node failures and recoveries
- Add Nodes for better performance
- Utilize virtualization for rapid deployment
Installing a Hadoop Cluster
- Understand the NameNode
- Understand the secondary NameNode
- Understand the Data Node
- Understand the JobTracker
- Understand the TaskTracker
Understanding the MapReduce Flow
- Map data
- Shuffle and sorting
- Reduce data
- Utilize the "Write Once, Read Many" approach
Reviewing Job Performance
- Interpret console output
- Navigate the JobTracker UI
- Use TaskTracker logs
Configuring Nodes
- Understand Hadoop property management
- Manage core properties
- Manage HDFS properties
- Manage MapReduce properties
- Manage worker properties
- Restrict job property changes
Supporting Federated & HA File Systems
- Restore NameNode services
- Protect NameNode metadata
- Use Federated NameNodes
- Understand the NameNode HA model
- Configure a Federated HDFS system
- Create defensive copies of NameNode metadata
Controlling Jobs and Resources
- Schedule jobs
- Understand the FairScheduler
- Orchestrate Workflows
Importing Legacy and Continuous Data
- Utilize Sqoop
- Use Flume
- Understand Hive, Impala, and HBase
Maintaining HDFS
- Check block integrity
- Balance data across nodes
- Use HDFS Safe Mode
- Address other HDFS systems
- Restrict Node additions
Installing Ecosystem Packages
- Install Pig
- Install Hive
- Review HBase requirements
Improving Hadoop Security
- Review authentication & authorization
- Analyze Hadoop's authorization model
- Understand kerberos architecture
- Review Kerberos implementation options