This product was designed to manage the world's largest data centers. Now we want to extend that scale and sense of security to the rest of the world. Why don't we price based on data volume like everyone else? Because that model will eat you alive with unanticipated charges, forcing us to become a billing company rather than a software company. The number of concurrent workers determines our data protection and data mover pricing.
Each worker performs a "job" and that job performs an action on your network mounts/shares. A single worker is capable of migrating/replicating/restoring more than 1B files/objects and petabytes of data. The same as our enterprise workers. You probably don't have a billion files in your network shares. So a single worker can easily handle 1-4 jobs very quickly, but one job at a time. That may be enough concurrency or you might decide to add more workers.