You can write the query result into HDFS by issue "insert into dst_tbl
<your query>". Would you mind sharing what's the end goal of your query? Do
you simply need the result written to HDFS? Or are you fetching data for an
app?
Thanks,
Alan
On Sun, Jan 19, 2014 at 3:11 AM, Antony wrote:
Hello there,
I have noticed that, especially when the result set is quite large, it
takes quite some time to send the results to the client which issued the
query. When I issue these queries in Hive, Hive is faster, as the results
are written to HDFS.
Is it possible to let Impala write the results to HDFS and thus skip
sending all results back to one node?
I am currently using impala-shell with -B and -o arguments.
Regards,
Antony
To unsubscribe from this group and stop receiving emails from it, send an
email to impala-user+unsubscribe@cloudera.org.
To unsubscribe from this group and stop receiving emails from it, send an email to impala-user+unsubscribe@cloudera.org.Hello there,
I have noticed that, especially when the result set is quite large, it
takes quite some time to send the results to the client which issued the
query. When I issue these queries in Hive, Hive is faster, as the results
are written to HDFS.
Is it possible to let Impala write the results to HDFS and thus skip
sending all results back to one node?
I am currently using impala-shell with -B and -o arguments.
Regards,
Antony
To unsubscribe from this group and stop receiving emails from it, send an
email to impala-user+unsubscribe@cloudera.org.
