[ https://issues.jboss.org/browse/TEIID-2884?page=com.atlassian.jira.plugin... ] Kylin Soong edited comment on TEIID-2884 at 8/3/15 5:30 AM: ------------------------------------------------------------ Amazon EMR is a platform that integrate all Hadoop ecosystem products to Amazon, for Big Data processing, analyzing, ETL, etc. For Hadoop ecosystem products like HBase, Hive, Spark, Impala, etc, we already have relevant translators, also Amazon EMR have supplied some compatiable driver, downloadable image, document, etc. http://docs.aws.amazon.com/ElasticMapReduce/latest/DeveloperGuide/HiveJDB... http://docs.aws.amazon.com/ElasticMapReduce/latest/DeveloperGuide/impala-... I think Teiid translator can wrok well against these products in Amazon EMR(Not test so far). Excepts integrate Hadoop ecosystem products, Amazon EMR add some more features, like EMR can use Amazon S3 to store input data, log files, and output data. For S3, I do wonder whether we need to develop a tranlator, what mainly stored in S3 is large picture files, vedio files and big text files, it seems meaningless to develop a JDBC based translator on top it. Further more, IMO, the work to integrate S3 to enterprise application should belong to ESB, actually there already have some implmentations: * Camel S3 Component(http://camel.apache.org/aws-s3.html) * Mulesoft S3 Connector(https://www.mulesoft.org/connectors/amazon-simple-storage-servi... * RSSBus/CData S3(https://www.rssbus.com/solutions/s3/) Any idea or advice in this issue? was (Author: kylin): Amazon EMR is a platform that integrate all Hadoop ecosystem products to Amazon, for Big Data processing, analyzing, ETL, etc. For Hadoop ecosystem products like HBase, Hive, Spark, Impala, etc, we already have relevant translatore, also Amazon EMR have supplied some compatiable driver, downloadable image, document, etc. http://docs.aws.amazon.com/ElasticMapReduce/latest/DeveloperGuide/HiveJDB... http://docs.aws.amazon.com/ElasticMapReduce/latest/DeveloperGuide/impala-... I think Teiid translator can wrok well against these products in Amazon EMR(Not test so far). Excepts integrate Hadoop ecosystem products, Amazon EMR add some more features, like EMR can use Amazon S3 to store input data, log files, and output data. For S3, I do wonder whether we need to develop a tranlator, what mainly stored in S3 is large picture files, vedio files and big text files, it seems meaningless to develop a JDBC based translator on top it. Further more, IMO, the work to integrate S3 to enterprise application should belong to ESB, actually there already have some implmentations: * Camel S3 Component(http://camel.apache.org/aws-s3.html) * Mulesoft S3 Connector(https://www.mulesoft.org/connectors/amazon-simple-storage-servi... * RSSBus/CData S3(https://www.rssbus.com/solutions/s3/) Any idea or advice in this issue? -- This message was sent by Atlassian JIRA (v6.3.15#6346)