In a scenario of uneven data distribution, implementing the optimization strategy of Data Shuffling is essential. Data Shuffling redistributes data across the cluster to achieve a more balanced workload, preventing hotspots and ensuring efficient parallel processing in a Hadoop cluster.

In the context of handling memory-intensive applications, garbage collection is crucial in advanced Hadoop tuning. Efficient garbage collection helps reclaim memory occupied by unused objects, preventing memory leaks and enhancing the overall performance of Hadoop applications.

In a Hadoop cluster, the ResourceManager is typically scaled first for performance enhancement. It manages resource allocation and scheduling, and scaling it ensures better coordination of resources, leading to improved job execution and overall cluster performance.

Load Balancing is key to maintaining the efficiency of a Hadoop cluster as data volume grows. It ensures that the computational load is evenly distributed among the nodes in the cluster, preventing any single node from becoming a bottleneck.

In complex data transformation tasks involving multiple data sources, Apache Sqoop is a preferred approach. Sqoop facilitates efficient and accurate data transfer between Hadoop and relational databases, ensuring seamless integration of diverse data sources for comprehensive transformations.

The integration of Apache Pig with Apache Spark allows for enhanced data processing and analysis in Hadoop. Apache Spark provides in-memory processing and advanced analytics capabilities, complementing Pig's data processing capabilities and enabling more sophisticated data workflows.

Data Replication is the process in Hadoop that ensures no data loss in case of a DataNode failure. Hadoop replicates data across multiple DataNodes, and if one node fails, the replicated data on other nodes can be used, preventing data loss.

Oozie's bundling feature in complex data pipelines enhances workflow management by consolidating multiple workflows into a single unit. This simplifies coordination and execution, making it easier to manage and monitor complex data processing tasks efficiently.

The command used to import data from a relational database to HDFS in Sqoop is 'sqoop import.' This command initiates the process of transferring data from a source database to the Hadoop ecosystem for further analysis and processing.

DBInputFormat is ideal for processing structured data stored in databases in Hadoop. It allows Hadoop MapReduce jobs to read data from relational database tables, providing a convenient way to integrate Hadoop with structured data sources.

In Cascading, a 'Tap' represents an input source or output sink in the data processing pipeline. It serves as a connection to external data sources or destinations, allowing data to flow through the Cascading application for processing.

Replication is the process by which HDFS ensures that each data block has the correct number of replicas. This helps in achieving fault tolerance by storing multiple copies of data across different nodes in the cluster.