We had almost the same discussion on another thread. I know your article and I have read it but it does not provide more information than the official documenation. And it is not clear about concurrency issue with the embedded mode. Sure if there is only one user per embedded server, there won't be any issue. I don't want to build something then have to do advanced concurrency testing (which is hard to get it right) only to see that I have the same problem with this use of the server.
Regards Bertrand On Sun, Sep 30, 2012 at 7:11 PM, Dilip Joseph <dilip.antony.jos...@gmail.com > wrote: > The data store does is not embedded. I have described how I accessed hive > from jython at > http://csgrad.blogspot.in/2010/04/to-use-language-other-than-java-say.html. > The example there may be relevant for your use case. > > Dilip > > > On Sun, Sep 30, 2012 at 8:05 AM, Bertrand Dechoux <decho...@gmail.com>wrote: > >> Hi, >> >> As the thrift sever has concurrency issue, I need to find an alternative. >> I found out about the embedded mode for the jdbc client. >> https://cwiki.apache.org/Hive/hiveclient.html >> >> However, I am not sure to understand everything. >> * Does that mean the datastore should be embedded too? I don't think so >> but would like a confirmation. >> * What is exactly an embedded server? Does it simply run the server code >> synchronically like any code? Would it be safe to share the same jdbc >> driver among several threads? What if distinct instances are used? Should >> the embedded server considered as a local standalone server? >> >> Regards >> >> Bertrand Dechoux >> > > > > -- > _________________________________________ > Dilip Antony Joseph > http://csgrad.blogspot.com > http://www.marydilip.info > -- Bertrand Dechoux