Akka Use case discussion

Hi,
I was exploring akka for one of the use-case where I need to shard and persist data.
I want to build system which can easily scale up-to millions on entries with operations like:
a) put - stores entry somewhere
b) get(condition) : scan data wrt conditions
As scanning of some external database would be very costly, I was sort of thinking of having a custom system with sharding (to reduce scanning) and replication with on-disk persistence.
Honestly, It’s like creating a custom database which supports my use-case.
I don’t need strict consistency.
Would akka fit my usecase?

Thanks,
Abhi