Tuning YARN resource allocation parameters impacts the performance of a Hadoop cluster by optimizing resource utilization. Proper allocation ensures efficient task execution, maximizes parallelism, and minimizes resource contention, leading to improved overall cluster performance.

Hive's HQL (Hive Query Language) feature allows for the execution of MapReduce jobs with SQL-like queries. It provides a higher-level abstraction for processing data stored in Hadoop Distributed File System (HDFS) using familiar SQL syntax.

Hive uses Predicate Pushdown to optimize query performance by pushing the filtering conditions closer to the data source, reducing the amount of data scanned during a query and improving overall efficiency.

The primary tool used for monitoring Hadoop cluster performance is Hadoop ResourceManager. It provides information about the resource utilization, job execution, and overall health of the cluster. Administrators use ResourceManager to ensure efficient resource allocation and identify any performance bottlenecks.

When dealing with time-series data in Hadoop, the optimal combination for performance is using the Parquet file format with Snappy compression. Parquet is columnar storage, and Snappy provides fast compression, making it efficient for analytical queries on time-series data.

For optimal results in aggregating data from multiple sources with Hadoop Streaming API, the approach should involve using Combiners for Intermediate Aggregation. Combiners help reduce the amount of data transferred between mappers and reducers, improving overall performance in the aggregation process.

Sqoop can be integrated with Shell scripts for custom data handling during import/export processes. This allows users to execute custom logic or transformations on the data as it is moved between Hadoop and relational databases.

Partitioning data in HDFS improves processing efficiency by enhancing data locality. This means that computation is performed on nodes where the data is already stored, reducing the need for extensive data movement across the network and thereby improving overall processing speed.

Snapshot recovery techniques in Hadoop allow for the restoration of data to a specific point in time. Snapshots capture the state of the HDFS at a particular moment, providing a reliable way to recover data to a known and consistent state.

Apache Hive is primarily used for building data pipelines involving SQL-like queries in the Hadoop ecosystem. It provides a high-level query language, HiveQL, that allows users to express queries in a SQL-like syntax, making it easier for SQL users to work with Hadoop data.