Hi chamila, In hive you can't delete the column family, it's not supported in the cassandra strorage handler.
If your requirement is incrementally processing, then better option would be using 'Incremental Analysis' feature [1], which was introduced in BAM 2.4.0. This wasn't fully tested with external hadoop node, etc in BAM 2.4.0 and it will be properly tested in the next BAM release 2.4.1. But it was working well without any issue in the standalone setup in BAM 2.4.0, hence you could use that feature in IMO for your requirement. [1] http://docs.wso2.org/pages/viewpage.action?pageId=32345660 Thanks, Sinthuja. On Mon, Feb 3, 2014 at 11:49 PM, Chamila Wijayarathna <[email protected]>wrote: > Hi Sinthuja, > In our problem, our main requirement is to delete older data from > cassandra column family to reduce execution time of hive scripts. As we > found on internet [1] hive does not support operations like delete and > update. That's why we thought about archiving data from cassandra. Is there > any way we can do this through hive? > > 1. https://cwiki.apache.org/confluence/display/Hive/Tutorial > > Thank You. > > > On Mon, Feb 3, 2014 at 3:54 PM, Sinthuja Ragendran <[email protected]>wrote: > >> Hi chamila, >> >> BAM Cassandra archival feature is designed for event streams, not for all >> cassandra column family. Hence for arbitrary column family you can't use >> this feature. CEP and BAM uses different stream definition store, hence the >> streams defined in CEP is not accessible in BAM. And also there is a >> limitation in BAM archival feature, that you can't archive hours old data >> and it will handled with days granularity. But this can be fixed easily, we >> will fix it so that you can archive with hours granularity in the next >> release. >> >> Anyhow finally the archive feature is generating hive script based on the >> stream name and the date/time you have provided in the UI and executes in a >> scheduled time interval to archive the data from the source column family. >> Hence you can also write similar hive script in your own which reads from >> your original source column family and archive to another column family, >> which does the same job as archival feature. >> >> Thanks, >> Sinthuja. >> >> >> On Sun, Feb 2, 2014 at 11:00 PM, Chamila Wijayarathna >> <[email protected]>wrote: >> >>> Hello all, >>> I'm writing data into cassandra column family in BAM using CEP Cassandra >>> output adapter. >>> I need to send old data in this column family to an archive as mentioned >>> in [1]. >>> What should I use as stream name for this? In BAM management console it >>> doesn't show any event streams. I tried output event stream name I used in >>> CEP, but seems like BAM doesn't recognize it. >>> Also can I Archive data which are older like 4 hours or 6 hours? It >>> seems like BAM only support to archive data older than an integer number of >>> days. >>> >>> Thank You >>> >>> >>> 1. http://docs.wso2.org/display/BAM240/Archive+Cassandra+Data >>> -- >>> *Chamila Wijayarathna* >>> Engineering Intern, >>> WSO2 Inc. >>> >>> >>> _______________________________________________ >>> Dev mailing list >>> [email protected] >>> http://wso2.org/cgi-bin/mailman/listinfo/dev >>> >>> >> >> >> -- >> *Sinthuja Rajendran* >> Software Engineer <http://wso2.com/> >> WSO2, Inc.:http://wso2.com >> >> Blog: http://sinthu-rajan.blogspot.com/ >> Mobile: +94774273955 >> >> >> > > > -- > *Chamila Wijayarathna* > Engineering Intern, > WSO2 Inc. > > -- *Sinthuja Rajendran* Software Engineer <http://wso2.com/> WSO2, Inc.:http://wso2.com Blog: http://sinthu-rajan.blogspot.com/ Mobile: +94774273955
_______________________________________________ Dev mailing list [email protected] http://wso2.org/cgi-bin/mailman/listinfo/dev
