Clickhouse is the best way to analyze 10GB sized json by far.
Latest bunch of features add near-native json support. Coupled with ability to add extracted columns make the whole process easy. It is fast, you can use familiar SQL syntax, not constrainted to RAM limits.
It is a bit hard if you want to iteratively process file line-by line or use advanced SQL. And you have one-time cost of writing schema. Apart from that, I can't think of any downsides.
Latest bunch of features add near-native json support. Coupled with ability to add extracted columns make the whole process easy. It is fast, you can use familiar SQL syntax, not constrainted to RAM limits.
It is a bit hard if you want to iteratively process file line-by line or use advanced SQL. And you have one-time cost of writing schema. Apart from that, I can't think of any downsides.
Edit: clarify a bit