http://www.datastax.com/support-forums/topic/read-huge-volume-of-data-from-cassandra Software, Support, and Training for Apache Cassandra en-US Tue, 21 May 2013 09:25:29 +0000 http://bbpress.org/?v=1.0.3 q http://www.datastax.com/support-forums/search.php http://www.datastax.com/support-forums/topic/read-huge-volume-of-data-from-cassandra#post-7444 Thu, 08 Nov 2012 18:06:52 +0000 alexliu 7444@http://www.datastax.com/support-forums/ <p>Cassandra-JDBC is different approach than CQL. CQL is preferred than Cassandra-JDBC if you can get CQL use for your need.</p> <p>You can use any other cassandra client to parallel reading from your client side. Cassandra is built for high read and write throughput for a distributed system.</p> <p>Hadoop mapredue do a parallel reading or writing also work for some case. </p> http://www.datastax.com/support-forums/topic/read-huge-volume-of-data-from-cassandra#post-7434 Wed, 07 Nov 2012 22:18:31 +0000 Pearly 7434@http://www.datastax.com/support-forums/ <p>As per the documentation, sstable2json converts the on-disk SSTable representation of a column family into a JSON formatted document. But I don't think we can add a filter condition to this. {like get all rows where columnA='something1' and columnB='something2'}. Appreciate your confirmation here. </p> <p>If the above doesn't help, my next option is to use "cassandra-jdbc". In your earlier response, you didn't seem to mention anything on cassandra-jdbc, but only on the standalone "CQL" query. Appreciate if you can confirm the usage of CQL with the Cassandra-JDBC? Is it a standard approach?<br /> Also I hope we need to Secondary indexes required if we need to perform the CQL query. </p> <p>Other than the above solutions I mentioned, are there are any Standard approach If I want to read high volume of data from a Column Family based on some filter conditions? </p> <p>Appreciate you help in these, </p> http://www.datastax.com/support-forums/topic/read-huge-volume-of-data-from-cassandra#post-6159 Mon, 27 Aug 2012 08:04:36 +0000 alexliu 6159@http://www.datastax.com/support-forums/ <p>CQL is preferred for cassandra query. You can parallel read multiple CFs with load balance cross the cassandra ring. </p> <p>There is another tool sstable2json for bulk reading</p> <p><a href="http://www.datastax.com/docs/1.0/references/sstable2json" rel="nofollow">http://www.datastax.com/docs/1.0/references/sstable2json</a></p> <p>Cassandra comes hadroop integration including hive and pig, it implements hadoop API. You shouldn't worry about how the data is read. </p> http://www.datastax.com/support-forums/topic/read-huge-volume-of-data-from-cassandra#post-6111 Mon, 20 Aug 2012 23:49:35 +0000 Pearly 6111@http://www.datastax.com/support-forums/ <p>What is the preferred/best approach if I want to read huge volume of data from Cassandra for reporting purpose? Any difference in approach if the read is from one CF Vs multiple CFs. (my current need is from one CF)</p> <p>Is it optimal to use, cassandra-jdbc and CQL with "select x,y,z…..where /some/ condition"? Or does this problem brings up the need of Hive or PIG? </p> <p>To give a short description of my current environment, I have 2 DCs with 3 nodes on each DC. with nodes started with -s (for solar) </p>