Define workflow and sample code to grep 500k terms on 0.5GB text with AWS

I have two files. A) file with 500k terms. B) file with 0.5 GB raw text. I want to return matching lines in B which contain A. Max 50 lines. desired output: term 1: line A; term 1: line B; term 2: line C, etc. I am familiar with EC2, but I am hoping there's an efficient way to set this up (map reduce?)