Resource limitations in Hadoop jobs can vary based on the specific workload, with factors like CPU, network, and disk playing different roles. Compression can save space but may increase CPU usage during decompression. Hadoop is designed to gracefully handle failures of individual data nodes, ensuring no data loss and seamless task reassignment, while also managing network partitions effectively.
