ETL [...] typically involves reading large volumes of data from a database [...]
I don't think that's right: it seems to me that ETL first and foremost means (always has): getting data from outside to inside a database. The TRANSFORM is only necessary if the db (-constraint(s)) demand it, and there, obviously, perl can come in handy.
... and processing it in massively parallel fashion
"massively parallel"? That doesn't have much to do with ETL, does it? If the database can slurp data parallel (multicore, or multi-whatever), that's nice, but that doesn't seem very related to any ETL-job.
(btw, your first link 'Extract, transform, and load...' links to the node itself; I suppose that was a mistake? You had wikipedia ETL in mind, perhaps? )