To address the 'Small Files Problem,' a Hadoop administrator can use CombineFileInputFormat. This technique allows the efficient processing of small files by combining them into larger input splits, reducing the overhead associated with managing numerous small files and improving overall processing efficiency.

Capacity Planning is used to estimate the processing capacity required for a Hadoop cluster based on data processing needs. It involves analyzing factors like data volume, processing speed, and storage requirements to ensure optimal cluster performance.

Hadoop's ResourceManager is responsible for resource allocation and management in the cluster. It assists in monitoring cluster performance by efficiently allocating resources to applications, ensuring optimal utilization and performance. This includes managing memory, CPU, and other resources for running tasks.

Apache Spark performs computations in-memory, which is a key improvement over the MapReduce model. This in-memory processing reduces the need for intermediate disk storage, resulting in faster data processing and analysis.

Impala's in-memory processing feature enables it to store and analyze data in memory, providing faster query performance and real-time data analysis capabilities in Hadoop clusters.

Data Serialization is a critical factor in Hadoop Streaming API when dealing with streaming data from various sources. Proper serialization ensures that the data is efficiently encoded and decoded, enhancing the performance of data processing in Hadoop Streaming.

Apache Flume facilitates building data pipelines in Hadoop by serving as a reliable and scalable data ingestion tool. It efficiently collects, aggregates, and moves large amounts of log data from various sources to Hadoop storage, making it a valuable component in data pipeline construction.

Rack awareness in Hadoop refers to the ability of the cluster to be aware of the physical location of nodes within a rack. It contributes to efficiency by optimizing data locality, ensuring that data processing is performed on nodes that are close to the stored data. This minimizes data transfer across the network, improving performance.

The primary role of DataNodes in a Hadoop cluster is to store and manage data blocks. They are responsible for storing the actual data and are distributed across the cluster to ensure fault tolerance and parallel data processing. DataNodes report to the NameNode about the health and status of the data blocks they store.

Split-by in Sqoop specifies the database column to be used for splitting the data during import. This is particularly useful when dealing with large datasets, allowing for parallel processing and efficient data import.