)]}'
{"specs/mitaka/hbase-in-trove.rst":[{"author":{"_account_id":17360,"name":"Ishita Mandhan","email":"imandha@us.ibm.com","username":"imandhan"},"change_message_id":"611729267c9c7aaa34fdae61aea097c8dd160909","unresolved":false,"context_lines":[{"line_number":79,"context_line":"In a fully distributed HBase installation, multiple HBase"},{"line_number":80,"context_line":"daemons run on multiple servers in a cluster. Data are stored in a"},{"line_number":81,"context_line":"fully distributed HDFS file system. The terminology is analogous to"},{"line_number":82,"context_line":"the fully distributed Hadoop Cluster."},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"Pseudo Distributed Mode"},{"line_number":85,"context_line":"-----------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a7e1126_02baa1c0","line":82,"range":{"start_line":82,"start_character":36,"end_line":82,"end_character":37},"updated":"2015-12-18 19:49:56.000000000","message":"what will be the data nodes that together make the hdfs?","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":17360,"name":"Ishita Mandhan","email":"imandha@us.ibm.com","username":"imandhan"},"change_message_id":"ee4c10758d86baf8139c5fc5c946e934103775ad","unresolved":false,"context_lines":[{"line_number":79,"context_line":"In a fully distributed HBase installation, multiple HBase"},{"line_number":80,"context_line":"daemons run on multiple servers in a cluster. Data are stored in a"},{"line_number":81,"context_line":"fully distributed HDFS file system. The terminology is analogous to"},{"line_number":82,"context_line":"the fully distributed Hadoop Cluster."},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"Pseudo Distributed Mode"},{"line_number":85,"context_line":"-----------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"fa69d971_91939ba3","line":82,"range":{"start_line":82,"start_character":36,"end_line":82,"end_character":37},"in_reply_to":"1a7b0d38_2768e5b8","updated":"2015-12-29 20:51:45.000000000","message":"Yup, that answered my question.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"d1e27f374ed4bcb6aec6cea2057b11acd8149fcf","unresolved":false,"context_lines":[{"line_number":79,"context_line":"In a fully distributed HBase installation, multiple HBase"},{"line_number":80,"context_line":"daemons run on multiple servers in a cluster. Data are stored in a"},{"line_number":81,"context_line":"fully distributed HDFS file system. The terminology is analogous to"},{"line_number":82,"context_line":"the fully distributed Hadoop Cluster."},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"Pseudo Distributed Mode"},{"line_number":85,"context_line":"-----------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"1a7b0d38_2768e5b8","line":82,"range":{"start_line":82,"start_character":36,"end_line":82,"end_character":37},"in_reply_to":"3a7e1126_02baa1c0","updated":"2015-12-28 02:35:42.000000000","message":"Ishita, in the fully distributed mode, they will be nova instances launched by Trove. I\u0027m not entirely sure I understand the question, I hope this answers your question.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":10295,"name":"mariam john","email":"mariamj@us.ibm.com","username":"mariamj"},"change_message_id":"d18f114fde1f7bc2d3bdc11b8ee62e1f4067c37c","unresolved":false,"context_lines":[{"line_number":114,"context_line":"| Python Trove Client (CLI)   |     X      |      X     |      X     |"},{"line_number":115,"context_line":"| Horizon Interface           |            |      X     |      X     |"},{"line_number":116,"context_line":"+-----------------------------+------------+------------+------------+"},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"It is tentatively proposed that Phase 1 will be in the Mitaka release"},{"line_number":119,"context_line":"timeframe, Phase II will be in the \u0027N\u0027 release and Phase III will be"},{"line_number":120,"context_line":"in the \u0027O\u0027 release."}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a7e1126_972ea341","line":117,"updated":"2015-12-18 15:33:33.000000000","message":"is the user/database management feature going to be included in any of these phases?","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"d1e27f374ed4bcb6aec6cea2057b11acd8149fcf","unresolved":false,"context_lines":[{"line_number":114,"context_line":"| Python Trove Client (CLI)   |     X      |      X     |      X     |"},{"line_number":115,"context_line":"| Horizon Interface           |            |      X     |      X     |"},{"line_number":116,"context_line":"+-----------------------------+------------+------------+------------+"},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"It is tentatively proposed that Phase 1 will be in the Mitaka release"},{"line_number":119,"context_line":"timeframe, Phase II will be in the \u0027N\u0027 release and Phase III will be"},{"line_number":120,"context_line":"in the \u0027O\u0027 release."}],"source_content_type":"text/x-rst","patch_set":1,"id":"1a7b0d38_879d19be","line":117,"in_reply_to":"3a7e1126_972ea341","updated":"2015-12-28 02:35:42.000000000","message":"Mariam, thanks for pointing that out. I will update the table. I believe that there is a concept of users but not of databases.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":10295,"name":"mariam john","email":"mariamj@us.ibm.com","username":"mariamj"},"change_message_id":"d18f114fde1f7bc2d3bdc11b8ee62e1f4067c37c","unresolved":false,"context_lines":[{"line_number":125,"context_line":"In the first phase the following will be submitted."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"1. Elements to create a guest image for HBase Standalone mode"},{"line_number":128,"context_line":"2. A guest agent for HBase Standalone mode"},{"line_number":129,"context_line":"3. Changes required in the CLI and the controller"},{"line_number":130,"context_line":"4. Unit tests"},{"line_number":131,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a7e1126_570a7b7a","line":128,"range":{"start_line":128,"start_character":5,"end_line":128,"end_character":10},"updated":"2015-12-18 15:33:33.000000000","message":"Is the guest image and agent going to be different for stand-alone vs psuedo distributed. The Work Item section also includes testing standalone and pseudo distributed mode but here you mention only standalone specifically.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"d1e27f374ed4bcb6aec6cea2057b11acd8149fcf","unresolved":false,"context_lines":[{"line_number":125,"context_line":"In the first phase the following will be submitted."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"1. Elements to create a guest image for HBase Standalone mode"},{"line_number":128,"context_line":"2. A guest agent for HBase Standalone mode"},{"line_number":129,"context_line":"3. Changes required in the CLI and the controller"},{"line_number":130,"context_line":"4. Unit tests"},{"line_number":131,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"1a7b0d38_5a470477","line":128,"range":{"start_line":128,"start_character":5,"end_line":128,"end_character":10},"in_reply_to":"3a7e1126_570a7b7a","updated":"2015-12-28 02:35:42.000000000","message":"Mariam, the spec (as you reviewed) was somewhat vague because I was still figuring out how best to implement the feature. Now that a v1 of the code is ready, I have a better idea and have updated the spec. You\u0027ll see impacts of that change throughout the new version I\u0027ll submit.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":17360,"name":"Ishita Mandhan","email":"imandha@us.ibm.com","username":"imandhan"},"change_message_id":"611729267c9c7aaa34fdae61aea097c8dd160909","unresolved":false,"context_lines":[{"line_number":152,"context_line":"identical to a current single node Trove database of any other type"},{"line_number":153,"context_line":"(like MySQL). The guest agent and the HBase instance will be on the"},{"line_number":154,"context_line":"same VM and the user will have access to HBase but not to ssh on the"},{"line_number":155,"context_line":"instance."},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"Architecture Overview for trove and Fully Distributed HBase"},{"line_number":158,"context_line":"-----------------------------------------------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a7e1126_a3911a49","line":155,"range":{"start_line":155,"start_character":8,"end_line":155,"end_character":9},"updated":"2015-12-18 19:49:56.000000000","message":"Maybe you could mention what will be different about the storage in this mode?","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":17360,"name":"Ishita Mandhan","email":"imandha@us.ibm.com","username":"imandhan"},"change_message_id":"ee4c10758d86baf8139c5fc5c946e934103775ad","unresolved":false,"context_lines":[{"line_number":152,"context_line":"identical to a current single node Trove database of any other type"},{"line_number":153,"context_line":"(like MySQL). The guest agent and the HBase instance will be on the"},{"line_number":154,"context_line":"same VM and the user will have access to HBase but not to ssh on the"},{"line_number":155,"context_line":"instance."},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"Architecture Overview for trove and Fully Distributed HBase"},{"line_number":158,"context_line":"-----------------------------------------------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"fa69d971_f1cf77b1","line":155,"range":{"start_line":155,"start_character":8,"end_line":155,"end_character":9},"in_reply_to":"1a7b0d38_fa603003","updated":"2015-12-29 20:51:45.000000000","message":"Right and in a fully distributed mode data is spread over nova instances as you answered in my above question.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"d1e27f374ed4bcb6aec6cea2057b11acd8149fcf","unresolved":false,"context_lines":[{"line_number":152,"context_line":"identical to a current single node Trove database of any other type"},{"line_number":153,"context_line":"(like MySQL). The guest agent and the HBase instance will be on the"},{"line_number":154,"context_line":"same VM and the user will have access to HBase but not to ssh on the"},{"line_number":155,"context_line":"instance."},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"Architecture Overview for trove and Fully Distributed HBase"},{"line_number":158,"context_line":"-----------------------------------------------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"1a7b0d38_fa603003","line":155,"range":{"start_line":155,"start_character":8,"end_line":155,"end_character":9},"in_reply_to":"3a7e1126_a3911a49","updated":"2015-12-28 02:35:42.000000000","message":"Ishita, in standalone and pseudo distributed mode, everything is on one nova instance. So the question is moot. I\u0027m not sure what the best way is for fully distributed mode.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":10295,"name":"mariam john","email":"mariamj@us.ibm.com","username":"mariamj"},"change_message_id":"d18f114fde1f7bc2d3bdc11b8ee62e1f4067c37c","unresolved":false,"context_lines":[{"line_number":256,"context_line":"   with all datastores is a very attractive one."},{"line_number":257,"context_line":""},{"line_number":258,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":259,"context_line":"   operational databases through a big-data data-lake solution, a"},{"line_number":260,"context_line":"   single abstraction of Trove is considered to be preferable to the"},{"line_number":261,"context_line":"   approach of having different datastores using different"},{"line_number":262,"context_line":"   abstractions."}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a7e1126_1770730a","line":259,"range":{"start_line":259,"start_character":49,"end_line":259,"end_character":53},"updated":"2015-12-18 15:33:33.000000000","message":"did you mean data-like instead of data-lake?","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"d1e27f374ed4bcb6aec6cea2057b11acd8149fcf","unresolved":false,"context_lines":[{"line_number":256,"context_line":"   with all datastores is a very attractive one."},{"line_number":257,"context_line":""},{"line_number":258,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":259,"context_line":"   operational databases through a big-data data-lake solution, a"},{"line_number":260,"context_line":"   single abstraction of Trove is considered to be preferable to the"},{"line_number":261,"context_line":"   approach of having different datastores using different"},{"line_number":262,"context_line":"   abstractions."}],"source_content_type":"text/x-rst","patch_set":1,"id":"1a7b0d38_9a380cf1","line":259,"range":{"start_line":259,"start_character":49,"end_line":259,"end_character":53},"in_reply_to":"3a7e1126_1770730a","updated":"2015-12-28 02:35:42.000000000","message":"Mariam, no. I was referring to data lake (https://en.wikipedia.org/wiki/Data_lake) which is NOT hyphenated. I will fix that.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":17360,"name":"Ishita Mandhan","email":"imandha@us.ibm.com","username":"imandhan"},"change_message_id":"611729267c9c7aaa34fdae61aea097c8dd160909","unresolved":false,"context_lines":[{"line_number":281,"context_line":"Work Items"},{"line_number":282,"context_line":"----------"},{"line_number":283,"context_line":""},{"line_number":284,"context_line":"For Mitaka-1, the work items are as follows."},{"line_number":285,"context_line":""},{"line_number":286,"context_line":"1. Build guest image elements"},{"line_number":287,"context_line":"2. Build guest image"}],"source_content_type":"text/x-rst","patch_set":1,"id":"7a740942_44b0ae43","line":284,"updated":"2015-12-18 19:49:56.000000000","message":"Mitaka-1?","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":17360,"name":"Ishita Mandhan","email":"imandha@us.ibm.com","username":"imandhan"},"change_message_id":"ee4c10758d86baf8139c5fc5c946e934103775ad","unresolved":false,"context_lines":[{"line_number":281,"context_line":"Work Items"},{"line_number":282,"context_line":"----------"},{"line_number":283,"context_line":""},{"line_number":284,"context_line":"For Mitaka-1, the work items are as follows."},{"line_number":285,"context_line":""},{"line_number":286,"context_line":"1. Build guest image elements"},{"line_number":287,"context_line":"2. Build guest image"}],"source_content_type":"text/x-rst","patch_set":1,"id":"fa69d971_91487b2f","line":284,"in_reply_to":"1a7b0d38_3a39d8f2","updated":"2015-12-29 20:51:45.000000000","message":"I meant mitaka-1 is over, it was Dec 1-3. Did you mean to write mitaka and not mitaka-1 specifically?","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"d1e27f374ed4bcb6aec6cea2057b11acd8149fcf","unresolved":false,"context_lines":[{"line_number":281,"context_line":"Work Items"},{"line_number":282,"context_line":"----------"},{"line_number":283,"context_line":""},{"line_number":284,"context_line":"For Mitaka-1, the work items are as follows."},{"line_number":285,"context_line":""},{"line_number":286,"context_line":"1. Build guest image elements"},{"line_number":287,"context_line":"2. Build guest image"}],"source_content_type":"text/x-rst","patch_set":1,"id":"1a7b0d38_3a39d8f2","line":284,"in_reply_to":"7a740942_44b0ae43","updated":"2015-12-28 02:35:42.000000000","message":"Mitaka-1 is the first drop for Mitaka. Is that the question?","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"1f5dba38dab9c14bf15e38b5ad27f73fb6258d90","unresolved":false,"context_lines":[{"line_number":281,"context_line":"Work Items"},{"line_number":282,"context_line":"----------"},{"line_number":283,"context_line":""},{"line_number":284,"context_line":"For Mitaka-1, the work items are as follows."},{"line_number":285,"context_line":""},{"line_number":286,"context_line":"1. Build guest image elements"},{"line_number":287,"context_line":"2. Build guest image"}],"source_content_type":"text/x-rst","patch_set":1,"id":"fa69d971_e05c7bef","line":284,"in_reply_to":"fa69d971_91487b2f","updated":"2015-12-30 03:53:18.000000000","message":"Yes, I will amend to Mitaka ... Sorry, didn\u0027t realize that.","commit_id":"9ab293c0996c498d433ed639b5fe809ca5a4e769"},{"author":{"_account_id":14823,"name":"Pete MacKinnon","email":"pmackinn@redhat.com","username":"pmackinn"},"change_message_id":"b0ad59d00ebd639b6a8191bc877e7d9c9f2daf92","unresolved":false,"context_lines":[{"line_number":428,"context_line":"   to manage a number of different database types and in the minds of"},{"line_number":429,"context_line":"   a lot of end-users, HBase is a database."},{"line_number":430,"context_line":""},{"line_number":431,"context_line":"2. Consider integration of Trove with Sahara."},{"line_number":432,"context_line":""},{"line_number":433,"context_line":"   This approach, while attractive has some clear"},{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_ade36c32","line":431,"updated":"2016-01-04 21:10:31.000000000","message":"Were there any discussions had with the Sahara team to explore this? I don\u0027t see any emails in the openstack-dev mailling list.","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"3df0098216633e6fe0d7b27807e70fc9b607de09","unresolved":false,"context_lines":[{"line_number":428,"context_line":"   to manage a number of different database types and in the minds of"},{"line_number":429,"context_line":"   a lot of end-users, HBase is a database."},{"line_number":430,"context_line":""},{"line_number":431,"context_line":"2. Consider integration of Trove with Sahara."},{"line_number":432,"context_line":""},{"line_number":433,"context_line":"   This approach, while attractive has some clear"},{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_56d6e0f1","line":431,"in_reply_to":"fa69d971_ade36c32","updated":"2016-01-05 14:19:45.000000000","message":"None that I know of.","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"},{"author":{"_account_id":14823,"name":"Pete MacKinnon","email":"pmackinn@redhat.com","username":"pmackinn"},"change_message_id":"b0ad59d00ebd639b6a8191bc877e7d9c9f2daf92","unresolved":false,"context_lines":[{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"},{"line_number":435,"context_line":"   establish a single big data and database management abstraction,"},{"line_number":436,"context_line":"   the value of a single API and a single mechanism for interacting"},{"line_number":437,"context_line":"   with all datastores is a very attractive one."},{"line_number":438,"context_line":""},{"line_number":439,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":440,"context_line":"   operational databases through a big-data data lake solution, a"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_4dde7012","line":437,"updated":"2016-01-04 21:10:31.000000000","message":"Unless that single API is not a good fit for what is essentially a SQL query processor and planner for Hadoop. Hadoop and HDFS is really the database but concepts like \"backup\" and \"resize\" don\u0027t really make sense in the domain of HDFS where data is inherently replicated.","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"},{"author":{"_account_id":10670,"name":"Michael McCune","email":"elmiko@redhat.com","username":"mimccune"},"change_message_id":"d9780bc4596c58dbf18bfcb189f35631e347d54d","unresolved":false,"context_lines":[{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"},{"line_number":435,"context_line":"   establish a single big data and database management abstraction,"},{"line_number":436,"context_line":"   the value of a single API and a single mechanism for interacting"},{"line_number":437,"context_line":"   with all datastores is a very attractive one."},{"line_number":438,"context_line":""},{"line_number":439,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":440,"context_line":"   operational databases through a big-data data lake solution, a"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_534d873c","line":437,"in_reply_to":"fa69d971_1d75e0f8","updated":"2016-01-05 17:20:44.000000000","message":"@Amrith, i\u0027m a little confused about the notion of a \"user who wishes to establish a single big data and database management abstraction\", what would be the single abstraction to big data *and* database management?","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"ef37f1aba3d4e4cc8073be908eb3ec13b216816a","unresolved":false,"context_lines":[{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"},{"line_number":435,"context_line":"   establish a single big data and database management abstraction,"},{"line_number":436,"context_line":"   the value of a single API and a single mechanism for interacting"},{"line_number":437,"context_line":"   with all datastores is a very attractive one."},{"line_number":438,"context_line":""},{"line_number":439,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":440,"context_line":"   operational databases through a big-data data lake solution, a"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_80d083f7","line":437,"in_reply_to":"fa69d971_1d75e0f8","updated":"2016-01-05 17:04:22.000000000","message":"Pete clarified in IRC that the issue of fit related to, for example, what does it mean to backup hbase. See http://eavesdrop.openstack.org/irclogs/%23openstack-trove/%23openstack-trove.2016-01-05.log 2016-01-05T16:55:41.\n\n2016-01-05T16:58:49  \u003cpmackinn\u003e amrith, e.g., what does it mean to backup Hbase?\n\n\n2016-01-05T16:59:26  \u003camrith\u003e it means, in Trove-speak, to serialize all the data in a form that can be used to launch a new instance at a future time.\n\n\n2016-01-05T16:59:37  \u003camrith\u003e and to store that on some inexpensive medium\n\n\n2016-01-05T16:59:41  \u003camrith\u003e such as swift\n\n\n2016-01-05T16:59:58  \u003camrith\u003e which is traditionally cheaper to operate than, say, cinder.\n\n\n2016-01-05T17:00:17  \u003camrith\u003e and the ability to do that in such a way that many such copies of a single instance can be made over time.\n\n\n2016-01-05T17:00:44  \u003camrith\u003e and in the future, extend that capability to fully-distributed hbase deployments with trove.\n\n\n2016-01-05T17:01:16  \u003camrith\u003e aka, same thing as it means for MySQL, PostgreSQL, MariaDB, MongoDB, Cassandra, CouchDB, Couchbase, Redis, ...","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"e19adb6e57d1e173891951192906be77c8a0d6db","unresolved":false,"context_lines":[{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"},{"line_number":435,"context_line":"   establish a single big data and database management abstraction,"},{"line_number":436,"context_line":"   the value of a single API and a single mechanism for interacting"},{"line_number":437,"context_line":"   with all datastores is a very attractive one."},{"line_number":438,"context_line":""},{"line_number":439,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":440,"context_line":"   operational databases through a big-data data lake solution, a"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_965d3849","line":437,"in_reply_to":"fa69d971_4dde7012","updated":"2016-01-05 14:24:26.000000000","message":"I\u0027m also a little bit at a loss to connect the dots between \"SQL query processor and planner for Hadoop\" with HBase.","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"3df0098216633e6fe0d7b27807e70fc9b607de09","unresolved":false,"context_lines":[{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"},{"line_number":435,"context_line":"   establish a single big data and database management abstraction,"},{"line_number":436,"context_line":"   the value of a single API and a single mechanism for interacting"},{"line_number":437,"context_line":"   with all datastores is a very attractive one."},{"line_number":438,"context_line":""},{"line_number":439,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":440,"context_line":"   operational databases through a big-data data lake solution, a"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_3647acbd","line":437,"in_reply_to":"fa69d971_4dde7012","updated":"2016-01-05 14:19:45.000000000","message":"That logic is equally applicable to a number of NoSQL datastores which can also deal with these things like replication and resize in different ways. See Cassandra Data Replication as an example of that.","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"},{"author":{"_account_id":9664,"name":"amrith","email":"amrith.kumar@gmail.com","username":"amrith"},"change_message_id":"cef95d4f05072b0e155a6670975debbb375cdf7a","unresolved":false,"context_lines":[{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"},{"line_number":435,"context_line":"   establish a single big data and database management abstraction,"},{"line_number":436,"context_line":"   the value of a single API and a single mechanism for interacting"},{"line_number":437,"context_line":"   with all datastores is a very attractive one."},{"line_number":438,"context_line":""},{"line_number":439,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":440,"context_line":"   operational databases through a big-data data lake solution, a"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_d1372f4b","line":437,"in_reply_to":"fa69d971_534d873c","updated":"2016-01-05 17:57:09.000000000","message":"Trove is solely in the management plane so NO! Trove does not intend to provide a data abstraction. Nor does this spec intend to make that transition.","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"},{"author":{"_account_id":14823,"name":"Pete MacKinnon","email":"pmackinn@redhat.com","username":"pmackinn"},"change_message_id":"98d37e04e68600751036f12722b61b986a34ae23","unresolved":false,"context_lines":[{"line_number":434,"context_line":"   downsides. Considering the perspective of an end user who wishes to"},{"line_number":435,"context_line":"   establish a single big data and database management abstraction,"},{"line_number":436,"context_line":"   the value of a single API and a single mechanism for interacting"},{"line_number":437,"context_line":"   with all datastores is a very attractive one."},{"line_number":438,"context_line":""},{"line_number":439,"context_line":"   To a user looking to implement an end-to-end solution from"},{"line_number":440,"context_line":"   operational databases through a big-data data lake solution, a"}],"source_content_type":"text/x-rst","patch_set":7,"id":"fa69d971_1d75e0f8","line":437,"in_reply_to":"fa69d971_965d3849","updated":"2016-01-05 16:48:03.000000000","message":"Apologies, got my Hive and Hbase confused. Nonetheless, the API fitness question remains.","commit_id":"b09b9bbb9c1e7a2fcf5c4044f154e9012f25b65d"}]}