How can DynomoDB Local support large datasets?

Question

I'm using DynomoDB Local to store and test against largish datasets, all is well up to around 5M records in a collection when it begins to get slow, too slow to even open and scan. At 15M records (7GB), it seems to lock up and become unresponsive, even after the scan times out.

The Java process is not using large memory, threads, etc. and the machine is more than powerful enough to load the same 15M records into a memory array but I need the data to persist.

Any ideas how to start DynomoDB local with more resources available?

Thanks

Answer

Please check below the points 
If possible, use queries instead of scans, and take advantage of indexes to retrieve data more efficiently.
Monitor the logs and metrics to see if there are any specific errors or bottlenecks that can be addressed.

How can DynomoDB Local support large datasets?

Relevant content