Is there something like push-down aggregation? (no replies)

Suppose, we have a really big table with billions of rows looking like this:

CREATE TABLE `foobar` (
`fooId` INT(10) UNSIGNED NOT NULL,
`barId` INT(10) UNSIGNED NOT NULL
);

Now, we want to run an aggregation function on that table, like:

SELECT fooId, COUNT(*) FROM foobar WHERE barId IN (<thousands of barId's>) GROUP BY fooId;


As far as I understand the MySQL cluster architecture, the data nodes are just storage nodes and 'rather dumb'. In the above case, they would send all relevant data (all rows with the barId's we're looking for) to the query node and the aggregation would happen there, thus causing a lot of network traffic and a lot of work for the query node.

Is it somehow possible to let the data node compute a partial aggregation (on its local data)? This would help a lot.

If it's not possible, are there any plans to implement a 'push-down aggregation' functionality?
And if not, why?