Database Reference
In-Depth Information
"dclocal_read_repair_chance" = "0.14",
"gc_grace_seconds" = "989898",
"bloom_filter_fp_chance" = "0.2",
"compaction" = "{'class' :
'LeveledCompactionStrategy'}",
"replicate_on_write" = "false", "caching" =
"all");
Please note, support for the Hive-managed Cassandra table is not yet available with
the Cassandra storage handler project. Trying to create a Hive-managed table as
CREATE TABLE twitter.twitterdata_hive(tweet_id string,
body string, tweeted_by string) STORED BY
'org.apache.hadoop.hive.cassandra.cql.CqlStorageHandler'
WITH SERDEPROPERTIES ("cql.primarykey" = "message_id,
author", "comment"="check", "read_repair_chance" = "0.2",
"dclocal_read_repair_chance" = "0.14", "gc_grace_seconds"
= "989898", "bloom_filter_fp_chance" = "0.2",
"compaction" = "{'class' : 'LeveledCompactionStrategy'}",
"replicate_on_write" = "false", "caching" = "all");
will result in the error shown in
Figure 6-21
.
Figure 6-21
.
shows error while creating internal table over Cassandra in Hive
Until now, we have explored various ways to perform MapReduce algorithms over
Cassandra via Hive and Pig. Another important aspect of data analytics is data migra-
tion, where existing RDBMS data needs to be migrated on Cassandra for scalability
and performance.
In the next section, we will explore more about data migration.
Data Migration
