Reduce import time of large volume data to elassandra

[seshagiri cherukuri]

Hello Everyone,

For one of our customer, we are importing data from a csv file to elassandra. The csv file will have 1500~ millions of records. currently this import process is taking 6~ hours.

we are trying to reduce this import time significantly.

We are running on single node cassandra, 7 core processor server. The table we are trying to import data is having 80 columns.

Could you please sugest us the right tuning parameters/best practices/right tools that you might have used.

we are using dsbulk to import csv data.

Regards,

Seshagiri.