It's not easy summarising it in one sentence, but 'auto batching' and 'caching' is what I meant by optimising the data retrieval. EDIT: the Haxl docs describe

It's not easy summarising it in one sentence, but 'auto batching' and 'caching' is what I meant by optimising the data retrieval. EDIT: the Haxl docs describe it as "a library and EDSL for efficient scheduling of concurrent data accesses with a concise applicative API"