What is a challenge of processing massive volumes of data that the MapReduce model needs to overcome?

Making sure our data processing application is fault tolerant

Merging together the results computed by multiple tasks executing in parallel on a cluster

Ensuring record boundaries are respected even when files are organized into separate blocks of fixed size

All the other choices are correct