Suggestion for improving cassandra performance

[Subodh Nijsure <su...@gmail.com>]

I have table with 128244 entries in it. I am running one cassandra
node on AWS EC2 x.large instance.
Cassandra is the only daemon running on this machine. Its SSD storage.

Its taking  cassandra python driver running on another machine 7
seconds to retrieve that data. This is pretty small table with
following entries:

                asset_id text,
                event_time timestamp,
                sensor_serial_number text,
                sensor_type int,
                temperature float,
                humidity float,
                polling_freq int,

I was wondering if there is anything I can do from cassandra
configuration side to make this faster?

I have already made the table with this option:
            AND caching = '{"keys":"ALL", "rows_per_partition":"ALL"}'

And in my cassandra.xml I have set row_cache_size_in_mb to  20. That
doesn't seem to make that much difference. Are there any other
parameters I could be tuning to cut down this fetch time?

-Subodh