cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From YES Linux <>
Subject importance of key cache vs row cache
Date Wed, 28 Jul 2010 19:26:05 GMT
i was wondering what the trade offs were between the key cache and row
cache?  which is more important from a read?  if you have a large row cache
can your key cache be small?

here is some background to my questions:

i have a data set that has alot of random access for rows using get slices
from a random partitioner.  we chose to use get slice to cut down on network
round trips and return all the data in one call.  we pull the full row back.
 the row usually less than 100 bytes in raw data minus over head from
cassandra.  we have alot of data that will be access from cassandra in
realtime and have very hard processing constraints, so we need to make sure
cassandra isn't the bottleneck for our growth.  we handle over 20 mill
requests per day that will access the cluster.


View raw message