Am 03.01.2015 um 07:07 schrieb Srinivasa T N: > Hi Wilm, > The reason is that for some auditing purpose, I want to store the > original files also. well, then I would use a hdfs cluster for storing, as it seems to be exactly what you need. If you collocate hdfs DataNodes and yarns ResourceManager, you also could spare a lot of hardware or costs for external services. It is not recommended to do that, but in your special case this should work. This seems applicable as you only use the hdfs for storing the xml exactly for that purpose.
But I'm more familiar with hadoop, hdfs and hbase than with Cassandra. So perhaps I'm biased. And what Jacob proposed could be a solution, too. Spares a lot of nerves ;). Best wishes, Wilm
