I run a cluster on EC2, at the moment I just dump tables to ctrl-A delimited flat files with INSERT OVERWRITE DIRECTORY and bulk load those to MySQL with LOAD DATA LOCAL INFILE:
something like this: LOAD DATA LOCAL INFILE '/mnt/pages.txt' INTO TABLE new_pages FIELDS TERMINATED BY 0x01 LINES TERMINATED BY '\n' (id, url, title, page_latest, total_pageviews, monthly_trend, daily_trend); There seems to be a bug where ctrl-A is the only delimiter used, regardless of the Hive table structure (tab delimited etc.) On Tue, Jul 14, 2009 at 10:45 AM, Edward Capriolo <[email protected]>wrote: > Min, > > Funny should should ask. I was talking to Aaron about sqoop, and > un-squoop (hadoop->SQL). I am also started building a dboutput udf > which I am about to open another thread about. It looks like we may > open up a hadoop Jira, and possible a Hive Jiira to chat about this. > Stay tuned! > > Edward > > On Tue, Jul 14, 2009 at 6:18 AM, Min Zhou<[email protected]> wrote: > > Hi all, > > > > How do you export hive tables into oracle/mysql? through oci(Oracle Call > > Interface), jdbc(oci/thin/MysqlDriver) or odbc? > > Written such tools in c/c++ through HDFS native library or in java(jdbc)? > > How do you translate hive tables' schema into oracle/mysql shema? > > > > Thanks, > > Min > > -- > > My research interests are distributed systems, parallel computing and > > bytecode based virtual machine. > > > > My profile: > > http://www.linkedin.com/in/coderplay > > My blog: > > http://coderplay.javaeye.com > > > -- Peter N. Skomoroch 617.285.8348 http://www.datawrangling.com http://delicious.com/pskomoroch http://twitter.com/peteskomoroch
