PayPal Hadoop Interview Questions
- Configure slots in Hadoop 2.0 and Hadoop 1.0.
- In case of high availability, if the connectivity between Standby and Active NameNode is lost. How will this impact the Hadoop cluster?
- What is the minimum number of ZooKeeper services required in Hadoop 2.0 and Hadoop 1.0?
- If the hardware quality of few machines in a Hadoop Cluster is very low. How will it affect the performance of the job and the overall performance of the cluster?
- How does a NameNode confirm that a particular node is dead?
- Explain the difference between blacklist node and dead node.
- How can you increase the NameNode heap memory?
- Configure capacity scheduler in Hadoop.
- After restarting the cluster, if the MapReduce jobs that were working earlier are failing now, what could have gone wrong while restarting?
- Explain the steps to add and remove a DataNode from the Hadoop cluster.
- In a large busy Hadoop cluster-how can you identify a long running job?
- When NameNode is down, what does the JobTracker do?
- When configuring Hadoop manually, which property file should be modified to configure slots?
- How will you add a new user to the cluster?
- What is the advantage of speculative execution? Under what situations, Speculative Execution might not be beneficial?