avro-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From sourabh chaki <chaki.sour...@gmail.com>
Subject How to process different types of avro schema
Date Mon, 18 Mar 2013 17:00:42 GMT
Hi All,****

In my application I am getting a stream of avro events. This stream
contains different types of avro events belonging to different schemas. I
was wondering what is the right way to process this data and do analytics
on top of this. Can I use hive? I did study the avro serde that could be
used to decode avro data and I’m thinking I need to transform the input
stream into (multiple) entries belonging to different tables. For this I’m
considering using a mapper job that would extract these events type by type
and then we could use hive on top of these separate schemas. I’m wondering
if anyone has dealt with such scenario before and if this approach would
work with decent performance?****

Alternative way is to use all the logic in M-R code for the analytics that
we want to do on top of this data. Please advise.****

Thanks in advance.****


View raw message