)]}'
{"specs/juno/big-data-sql-v2.rst":[{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"a6184c42588f993347aa6ee45607bf814b23e352","unresolved":false,"context_lines":[{"line_number":19,"context_line":"Problem description"},{"line_number":20,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"currently, the Metering data model is completely denormalise. we store"},{"line_number":23,"context_line":"Samples, which are the raw data points, and Meters, which are the definition"},{"line_number":24,"context_line":"of said data points. this optimisation allows for good write performance but"},{"line_number":25,"context_line":"due to size of Sample table, can cause issues with read performance"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_9061dd13","line":22,"updated":"2014-06-24 19:11:40.000000000","message":"Non-block nitpick (only fix if uploading a new patch anyway):\n\n  s/denormalise/denormalized/","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":3012,"name":"Dina Belova","email":"dbelova@mirantis.com","username":"dbelova"},"change_message_id":"cd8d4b796f337d4c897d718059892fcade0438f5","unresolved":false,"context_lines":[{"line_number":19,"context_line":"Problem description"},{"line_number":20,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"currently, the Metering data model is completely denormalise. we store"},{"line_number":23,"context_line":"Samples, which are the raw data points, and Meters, which are the definition"},{"line_number":24,"context_line":"of said data points. this optimisation allows for good write performance but"},{"line_number":25,"context_line":"due to size of Sample table, can cause issues with read performance"}],"source_content_type":"text/x-rst","patch_set":6,"id":"baada198_f8e354d5","line":22,"in_reply_to":"1ae5cdf2_9061dd13","updated":"2014-07-28 10:08:58.000000000","message":"we -\u003e We\n\nand over all the text new sentences do not start with the capital letters :)","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"df5d28ceebf7a5cdd98ff4d0fada3d761880f378","unresolved":false,"context_lines":[{"line_number":19,"context_line":"Problem description"},{"line_number":20,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"currently, the Metering data model is completely denormalise. we store"},{"line_number":23,"context_line":"Samples, which are the raw data points, and Meters, which are the definition"},{"line_number":24,"context_line":"of said data points. this optimisation allows for good write performance but"},{"line_number":25,"context_line":"due to size of Sample table, can cause issues with read performance"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_48578027","line":22,"in_reply_to":"1ae5cdf2_9061dd13","updated":"2014-06-25 19:26:51.000000000","message":"will do","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"03a8aec802890cf4d98798ce5d1850bee9f91c7f","unresolved":false,"context_lines":[{"line_number":67,"context_line":"\t  * name: meter name"},{"line_number":68,"context_line":"\t  * type: meter type"},{"line_number":69,"context_line":"\t  * unit: meter unit"},{"line_number":70,"context_line":"\t  * resource_id: resource uuid  (-\u003eresource.id)"},{"line_number":71,"context_line":"\t* sample - the raw incoming data"},{"line_number":72,"context_line":"\t  * id: sample id"},{"line_number":73,"context_line":"\t  * meter_id: meter id (-\u003emeter.id)"}],"source_content_type":"text/x-rst","patch_set":6,"id":"baada198_8c7f7bdf","line":70,"updated":"2014-07-07 15:38:34.000000000","message":"This implies that there is a row of meter definition per resource whereas in the artifact world there is a class of meter for which many resources want to record samples. Or to put it another way a sample can only be from one resource whereas as a meter can be associated with many resources and a resource can be associated with many meters.\n\nThis makes the proposed schema seem a bit awkward. Does there need to be an association between meter and resource directly or can that association exist solely as a composite foreign key in the sample? If not, can the why be explained here in the spec? Along the lines of \"we have to be able to make queries of the sort blah blah, as a result, blah blah\".\n\nIf I\u0027ve missed some piece of the puzzle that ought to have been obvious, please point it out. Thank you.","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"042936c616c42514c2e4bde14a3eeaaa9f24c2ab","unresolved":false,"context_lines":[{"line_number":67,"context_line":"\t  * name: meter name"},{"line_number":68,"context_line":"\t  * type: meter type"},{"line_number":69,"context_line":"\t  * unit: meter unit"},{"line_number":70,"context_line":"\t  * resource_id: resource uuid  (-\u003eresource.id)"},{"line_number":71,"context_line":"\t* sample - the raw incoming data"},{"line_number":72,"context_line":"\t  * id: sample id"},{"line_number":73,"context_line":"\t  * meter_id: meter id (-\u003emeter.id)"}],"source_content_type":"text/x-rst","patch_set":6,"id":"baada198_5746d4c0","line":70,"in_reply_to":"baada198_8c7f7bdf","updated":"2014-07-21 18:31:09.000000000","message":"good catch. stupid design on my part. so i think this should be switched. resource_id should be attached to sample and meter_id an attribute of sample and possibly resource? from what i remember from meeting Bayer recommended avoiding joins as much as possible","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"a6184c42588f993347aa6ee45607bf814b23e352","unresolved":false,"context_lines":[{"line_number":118,"context_line":"potential difference in metadata between samples"},{"line_number":119,"context_line":""},{"line_number":120,"context_line":"**NOTE: Up for discussion to be removed in final revision**"},{"line_number":121,"context_line":"should we continue to store a unique instance of metadata for each sample?"},{"line_number":122,"context_line":"* another option is to delay this decision until another patch which includes"},{"line_number":123,"context_line":"properly implementing metadata tables as EAV"},{"line_number":124,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_100a2d27","line":121,"updated":"2014-06-24 19:11:40.000000000","message":"For v2, yes I think so. But clearly removing this is the key motivator for v3.\n\nHowever I think we\u0027re stuck with it for v2, i.e. we can\u0027t remove it via \"death by a thousand cuts\". Rather it would have to go as part of a well-defined major switch-over to v3, with a long deprecation path for the \"classic\" logic.","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"df5d28ceebf7a5cdd98ff4d0fada3d761880f378","unresolved":false,"context_lines":[{"line_number":118,"context_line":"potential difference in metadata between samples"},{"line_number":119,"context_line":""},{"line_number":120,"context_line":"**NOTE: Up for discussion to be removed in final revision**"},{"line_number":121,"context_line":"should we continue to store a unique instance of metadata for each sample?"},{"line_number":122,"context_line":"* another option is to delay this decision until another patch which includes"},{"line_number":123,"context_line":"properly implementing metadata tables as EAV"},{"line_number":124,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_48a94024","line":121,"in_reply_to":"1ae5cdf2_100a2d27","updated":"2014-06-25 19:26:51.000000000","message":"cool. let\u0027s stick with having a unique copy of metadata per sample for now","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"a6184c42588f993347aa6ee45607bf814b23e352","unresolved":false,"context_lines":[{"line_number":120,"context_line":"**NOTE: Up for discussion to be removed in final revision**"},{"line_number":121,"context_line":"should we continue to store a unique instance of metadata for each sample?"},{"line_number":122,"context_line":"* another option is to delay this decision until another patch which includes"},{"line_number":123,"context_line":"properly implementing metadata tables as EAV"},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Performance/Scalability Impacts"},{"line_number":126,"context_line":"-------------------------------"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_70fff102","line":123,"updated":"2014-06-24 19:11:40.000000000","message":"A-ha, OK, a \"sql phase 3\" BP for juno-3?","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"df5d28ceebf7a5cdd98ff4d0fada3d761880f378","unresolved":false,"context_lines":[{"line_number":120,"context_line":"**NOTE: Up for discussion to be removed in final revision**"},{"line_number":121,"context_line":"should we continue to store a unique instance of metadata for each sample?"},{"line_number":122,"context_line":"* another option is to delay this decision until another patch which includes"},{"line_number":123,"context_line":"properly implementing metadata tables as EAV"},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Performance/Scalability Impacts"},{"line_number":126,"context_line":"-------------------------------"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_687fdcab","line":123,"in_reply_to":"1ae5cdf2_70fff102","updated":"2014-06-25 19:26:51.000000000","message":"yeah, i was debating whether doing EAV work should go first or this work. the EAV work i don\u0027t think really affects performance much. it really just affects metadata query and also increases db size due to redundancies.","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"a6184c42588f993347aa6ee45607bf814b23e352","unresolved":false,"context_lines":[{"line_number":130,"context_line":"the write performance is not expected to degrade noticeably."},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"it is expected any degradation in write performance would be caught by"},{"line_number":133,"context_line":"existing tempest tests"},{"line_number":134,"context_line":""},{"line_number":135,"context_line":""},{"line_number":136,"context_line":"Other deployer impact"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_eb0a341f","line":133,"updated":"2014-06-24 19:11:40.000000000","message":"And also by performance profiling in advance with Ilya\u0027s loadgen harness?","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"88eeb70af276d82dc87fd6e682116e630939c085","unresolved":false,"context_lines":[{"line_number":130,"context_line":"the write performance is not expected to degrade noticeably."},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"it is expected any degradation in write performance would be caught by"},{"line_number":133,"context_line":"existing tempest tests"},{"line_number":134,"context_line":""},{"line_number":135,"context_line":""},{"line_number":136,"context_line":"Other deployer impact"}],"source_content_type":"text/x-rst","patch_set":6,"id":"baada198_2a9f4b04","line":133,"in_reply_to":"1ae5cdf2_e8b4cc83","updated":"2014-07-06 20:32:54.000000000","message":"Ilya originally had a fork here:\n\n  https://github.com/ityaptin/ceilometer/blob/master/tools/sample-generator.py\n\nbut has gone ahead and proposed this back into the main repo via:\n\n  https://review.openstack.org/102225","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"df5d28ceebf7a5cdd98ff4d0fada3d761880f378","unresolved":false,"context_lines":[{"line_number":130,"context_line":"the write performance is not expected to degrade noticeably."},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"it is expected any degradation in write performance would be caught by"},{"line_number":133,"context_line":"existing tempest tests"},{"line_number":134,"context_line":""},{"line_number":135,"context_line":""},{"line_number":136,"context_line":"Other deployer impact"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_e8b4cc83","line":133,"in_reply_to":"1ae5cdf2_eb0a341f","updated":"2014-06-25 19:26:51.000000000","message":"i\u0027m unaware of loadgen work... happen to have a link? i\u0027ll add it to this bp so i remember to benchmark before/after","commit_id":"a217b747a9553bd853e80fc97490271585aebc78"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"165a215258325213825724ffe8ba2b074c65e364","unresolved":false,"context_lines":[{"line_number":29,"context_line":"The current schema is as follows::"},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"\t* meter - meter definition"},{"line_number":32,"context_line":"\t  * id: meter def id"},{"line_number":33,"context_line":"\t  * name: meter name"},{"line_number":34,"context_line":"\t  * type: meter type"},{"line_number":35,"context_line":"\t  * unit: meter unit"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_4d434813","line":32,"updated":"2014-07-30 08:01:42.000000000","message":"Clearer as s/meter def id/meter id/ ?","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"90da250dde4b8da5d22a60e859ed3cd44d11fcd3","unresolved":false,"context_lines":[{"line_number":29,"context_line":"The current schema is as follows::"},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"\t* meter - meter definition"},{"line_number":32,"context_line":"\t  * id: meter def id"},{"line_number":33,"context_line":"\t  * name: meter name"},{"line_number":34,"context_line":"\t  * type: meter type"},{"line_number":35,"context_line":"\t  * unit: meter unit"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_2b86bcba","line":32,"in_reply_to":"baada198_4d434813","updated":"2014-07-31 19:41:19.000000000","message":"Done","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"adf848802cb98a99f2e78809e09382c14f8b1848","unresolved":false,"context_lines":[{"line_number":56,"context_line":""},{"line_number":57,"context_line":"the proposed schema is as follows::"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"    * resource - resource data"},{"line_number":60,"context_line":"        * id: resource uuid"},{"line_number":61,"context_line":"        * user_id: user uuid"},{"line_number":62,"context_line":"        * project_id: project uuid"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_00c90e51","line":59,"updated":"2014-07-30 11:48:46.000000000","message":"Do we have hand-wavey numbers for how many resources there might be in an \"average\" deployment and how many of those resources will have N\u003emany  resources associated?\n\nJust so we can get a general idea on the number of nearly duplicate rows this table would have.\n\nIf it is significantly high it might make sense to normalize, depending on the types of join that will happen, or even just to remove meter_id entirely, and rely on the sample to provide both meter id and resource id.\n\nI guess this fits in with Gordon and Mike\u0027s comments to build some tables and EXPLAIN some queries. If we\u0027re going to do that I would hope we start from the least redundant schema and add redundancies as necessary.","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"90da250dde4b8da5d22a60e859ed3cd44d11fcd3","unresolved":false,"context_lines":[{"line_number":56,"context_line":""},{"line_number":57,"context_line":"the proposed schema is as follows::"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"    * resource - resource data"},{"line_number":60,"context_line":"        * id: resource uuid"},{"line_number":61,"context_line":"        * user_id: user uuid"},{"line_number":62,"context_line":"        * project_id: project uuid"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_ab0f2c8a","line":59,"in_reply_to":"baada198_00c90e51","updated":"2014-07-31 19:41:19.000000000","message":"this would be 1:1 with sample table","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"165a215258325213825724ffe8ba2b074c65e364","unresolved":false,"context_lines":[{"line_number":61,"context_line":"        * user_id: user uuid"},{"line_number":62,"context_line":"        * project_id: project uuid"},{"line_number":63,"context_line":"        * source_id: source id"},{"line_number":64,"context_line":"        * meter_id: meter id "},{"line_number":65,"context_line":"        * resource_metadata: metadata dictionary"},{"line_number":66,"context_line":"    * meter - meter definition"},{"line_number":67,"context_line":"        * id: meter def id"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_0dd5d01a","line":64,"updated":"2014-07-30 08:01:42.000000000","message":"Nit-pick: trailing whitespace","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":4491,"name":"Lianhao Lu","email":"llh_misc@outlook.com","username":"lianhao-lu"},"change_message_id":"9b58d84298137107143d771b551e322fa5512aac","unresolved":false,"context_lines":[{"line_number":61,"context_line":"        * user_id: user uuid"},{"line_number":62,"context_line":"        * project_id: project uuid"},{"line_number":63,"context_line":"        * source_id: source id"},{"line_number":64,"context_line":"        * meter_id: meter id "},{"line_number":65,"context_line":"        * resource_metadata: metadata dictionary"},{"line_number":66,"context_line":"    * meter - meter definition"},{"line_number":67,"context_line":"        * id: meter def id"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_c58d447e","line":64,"updated":"2014-07-31 06:41:43.000000000","message":"why meter_id here? how about a single resource with multiple meters? i.e. a VM instance has several meters, cpu/memory/etc. Should we add resource_id into the meter table instead?","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"90da250dde4b8da5d22a60e859ed3cd44d11fcd3","unresolved":false,"context_lines":[{"line_number":61,"context_line":"        * user_id: user uuid"},{"line_number":62,"context_line":"        * project_id: project uuid"},{"line_number":63,"context_line":"        * source_id: source id"},{"line_number":64,"context_line":"        * meter_id: meter id "},{"line_number":65,"context_line":"        * resource_metadata: metadata dictionary"},{"line_number":66,"context_line":"    * meter - meter definition"},{"line_number":67,"context_line":"        * id: meter def id"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_4b81309f","line":64,"in_reply_to":"baada198_0dd5d01a","updated":"2014-07-31 19:41:19.000000000","message":"Done","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"db134423e8abdae2d46c2f31b0ac97031c6980f9","unresolved":false,"context_lines":[{"line_number":61,"context_line":"        * user_id: user uuid"},{"line_number":62,"context_line":"        * project_id: project uuid"},{"line_number":63,"context_line":"        * source_id: source id"},{"line_number":64,"context_line":"        * meter_id: meter id "},{"line_number":65,"context_line":"        * resource_metadata: metadata dictionary"},{"line_number":66,"context_line":"    * meter - meter definition"},{"line_number":67,"context_line":"        * id: meter def id"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_d760880f","line":64,"in_reply_to":"baada198_c58d447e","updated":"2014-07-31 15:41:17.000000000","message":"so i actually did have that originally. the issue becomes we now have have 3 tables, all with n number of rows. (since we have a unique resource for every sample).\n\nhaving meter_id in resource i think we can build a backref to make the connection to meter table without increasing data size","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"165a215258325213825724ffe8ba2b074c65e364","unresolved":false,"context_lines":[{"line_number":64,"context_line":"        * meter_id: meter id "},{"line_number":65,"context_line":"        * resource_metadata: metadata dictionary"},{"line_number":66,"context_line":"    * meter - meter definition"},{"line_number":67,"context_line":"        * id: meter def id"},{"line_number":68,"context_line":"        * name: meter name"},{"line_number":69,"context_line":"        * type: meter type"},{"line_number":70,"context_line":"        * unit: meter unit"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_8de000fa","line":67,"updated":"2014-07-30 08:01:42.000000000","message":"Clearer as s/meter def id/meter id/ ?","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"90da250dde4b8da5d22a60e859ed3cd44d11fcd3","unresolved":false,"context_lines":[{"line_number":64,"context_line":"        * meter_id: meter id "},{"line_number":65,"context_line":"        * resource_metadata: metadata dictionary"},{"line_number":66,"context_line":"    * meter - meter definition"},{"line_number":67,"context_line":"        * id: meter def id"},{"line_number":68,"context_line":"        * name: meter name"},{"line_number":69,"context_line":"        * type: meter type"},{"line_number":70,"context_line":"        * unit: meter unit"}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_6b90b4ed","line":67,"in_reply_to":"baada198_8de000fa","updated":"2014-07-31 19:41:19.000000000","message":"Done","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":4491,"name":"Lianhao Lu","email":"llh_misc@outlook.com","username":"lianhao-lu"},"change_message_id":"9b58d84298137107143d771b551e322fa5512aac","unresolved":false,"context_lines":[{"line_number":90,"context_line":"described above:"},{"line_number":91,"context_line":""},{"line_number":92,"context_line":"* creating a new Resource table"},{"line_number":93,"context_line":"* adding resource_id to meter table"},{"line_number":94,"context_line":"* moving appropriate values from sample table to resource table"},{"line_number":95,"context_line":""},{"line_number":96,"context_line":"these changes will require database migration."}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_e5b6c824","line":93,"updated":"2014-07-31 06:41:43.000000000","message":"why add resource_id to meter table? I don\u0027t see resource_id field in the above schema.","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"db134423e8abdae2d46c2f31b0ac97031c6980f9","unresolved":false,"context_lines":[{"line_number":90,"context_line":"described above:"},{"line_number":91,"context_line":""},{"line_number":92,"context_line":"* creating a new Resource table"},{"line_number":93,"context_line":"* adding resource_id to meter table"},{"line_number":94,"context_line":"* moving appropriate values from sample table to resource table"},{"line_number":95,"context_line":""},{"line_number":96,"context_line":"these changes will require database migration."}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_37d2a4d1","line":93,"in_reply_to":"baada198_e5b6c824","updated":"2014-07-31 15:41:17.000000000","message":"whoops, didn\u0027t update","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"165a215258325213825724ffe8ba2b074c65e364","unresolved":false,"context_lines":[{"line_number":127,"context_line":"existing tempest tests."},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"Use of Ilya\u0027s performance tool will be used to verify there is negligible"},{"line_number":130,"context_line":"performance degradation.[1]"},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"[1] https://github.com/ityaptin/ceilometer/blob/master/tools/sample-generator.py"},{"line_number":133,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_6dcacc70","line":130,"updated":"2014-07-30 08:01:42.000000000","message":"s/negligible performance degradation/improved read performance and negligible write performance degradation/","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"90da250dde4b8da5d22a60e859ed3cd44d11fcd3","unresolved":false,"context_lines":[{"line_number":127,"context_line":"existing tempest tests."},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"Use of Ilya\u0027s performance tool will be used to verify there is negligible"},{"line_number":130,"context_line":"performance degradation.[1]"},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"[1] https://github.com/ityaptin/ceilometer/blob/master/tools/sample-generator.py"},{"line_number":133,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"baada198_8b8b48bf","line":130,"in_reply_to":"baada198_6dcacc70","updated":"2014-07-31 19:41:19.000000000","message":"Done","commit_id":"9a4beb01c7d61d014ca0b25f0702aa0053184c05"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"acd9873171cc036b0b0d73fc968cab97fea69b2c","unresolved":false,"context_lines":[{"line_number":57,"context_line":"the proposed schema is as follows::"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"    * resource - resource data"},{"line_number":60,"context_line":"        * id: resource uuid"},{"line_number":61,"context_line":"        * user_id: user uuid"},{"line_number":62,"context_line":"        * project_id: project uuid"},{"line_number":63,"context_line":"        * source_id: source id"}],"source_content_type":"text/x-rst","patch_set":9,"id":"baada198_3a652049","line":60,"updated":"2014-07-31 19:54:21.000000000","message":"actually i need to change this as well. i just realised that since we have a unique resource entry for each sample i\u0027ll need a unique identifier column for each entry since resource uuid is not unique...\n\nwe can also probably further normalise by pulling metadata into it\u0027s own place since that\u0027s what is changing but i will leave that to a future change if necessary.","commit_id":"3fbc46da75fd96a069fae72df935e08d77c08009"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"0067419c78eba03f6e261ef910de87df786f9e42","unresolved":false,"context_lines":[{"line_number":62,"context_line":"        * project_id: project uuid"},{"line_number":63,"context_line":"        * source_id: source id"},{"line_number":64,"context_line":"        * meter_id: meter id"},{"line_number":65,"context_line":"        * resource_metadata: metadata dictionary"},{"line_number":66,"context_line":"    * meter - meter definition"},{"line_number":67,"context_line":"        * id: meter id"},{"line_number":68,"context_line":"        * name: meter name"}],"source_content_type":"text/x-rst","patch_set":9,"id":"baada198_bf31220c","line":65,"updated":"2014-07-31 19:28:14.000000000","message":"If the idea is to implement a normali[zs]ed model I still don\u0027t understand why meter_id is in here?","commit_id":"3fbc46da75fd96a069fae72df935e08d77c08009"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"345099e54c4c1d951afb9c5fbdaa798de7cc01ac","unresolved":false,"context_lines":[{"line_number":62,"context_line":"        * project_id: project uuid"},{"line_number":63,"context_line":"        * source_id: source id"},{"line_number":64,"context_line":"        * meter_id: meter id"},{"line_number":65,"context_line":"        * resource_metadata: metadata dictionary"},{"line_number":66,"context_line":"    * meter - meter definition"},{"line_number":67,"context_line":"        * id: meter id"},{"line_number":68,"context_line":"        * name: meter name"}],"source_content_type":"text/x-rst","patch_set":9,"id":"baada198_7a979845","line":65,"in_reply_to":"baada198_bf31220c","updated":"2014-07-31 19:41:04.000000000","message":"not technically necessary but it\u0027s possible to use this relationship to build a backref to meter table so if when building meter model (https://github.com/openstack/ceilometer/blob/master/ceilometer/storage/impl_sqlalchemy.py#L447-L454)\n\nit\u0027s possible to be build meter model without using sample table as intermediary... that is the idea.. whether EXPLAIN/performance prove otherwise is another story.","commit_id":"3fbc46da75fd96a069fae72df935e08d77c08009"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"be80c70f5e8f3ec3160f5ec86288278c73fdc449","unresolved":false,"context_lines":[{"line_number":58,"context_line":""},{"line_number":59,"context_line":"    * resource - resource data"},{"line_number":60,"context_line":"        * internal_id: resource id (to handle assumption resources may not be"},{"line_number":61,"context_line":"                       unique ie. diff user/project/source/meta per resource)"},{"line_number":62,"context_line":"        * resource_id: resource uuid"},{"line_number":63,"context_line":"        * user_id: user uuid"},{"line_number":64,"context_line":"        * project_id: project uuid"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_2134fff9","line":61,"updated":"2014-08-08 19:57:03.000000000","message":"TBH as discussed on IRC I\u0027m not 100% convinced of the value of maintaining that ability to mutate the user/project/source for resources, as it doesn\u0027t reflect the reality of these being immutable.\n\nThat said, I\u0027m OK with being conservative and letting this slide until v3.","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"8f38f6163e2d61cdcf9769934d22b8c10bb60252","unresolved":false,"context_lines":[{"line_number":58,"context_line":""},{"line_number":59,"context_line":"    * resource - resource data"},{"line_number":60,"context_line":"        * internal_id: resource id (to handle assumption resources may not be"},{"line_number":61,"context_line":"                       unique ie. diff user/project/source/meta per resource)"},{"line_number":62,"context_line":"        * resource_id: resource uuid"},{"line_number":63,"context_line":"        * user_id: user uuid"},{"line_number":64,"context_line":"        * project_id: project uuid"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_a7e63bad","line":61,"in_reply_to":"baada198_2134fff9","updated":"2014-08-11 10:25:49.000000000","message":"If we were really to go whole hog with the stuff I\u0027d suggest separating the resource (id, user, project, source) and metadata into separate tables.\n\nBut as with everyone else it is hard to speculate on how much time or risk that would entail.\n\nI\u0027m experimenting with changing the tests so they reflect reality on this patchset: https://review.openstack.org/#/c/113014/","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"},{"author":{"_account_id":1669,"name":"Julien Danjou","display_name":"jd","email":"julien@danjou.info","username":"jdanjou"},"change_message_id":"949b3489189a91e19d1a8ca7c465edf46c2d35d2","unresolved":false,"context_lines":[{"line_number":83,"context_line":"------------"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":"As a schema can be defined in quite a few ways, there are many alternatives in"},{"line_number":86,"context_line":"that sense."},{"line_number":87,"context_line":""},{"line_number":88,"context_line":"Data model impact"},{"line_number":89,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_ecf0ec93","line":86,"updated":"2014-08-12 09:49:32.000000000","message":"True that, we could also lay on the beach and do nothing. ;)","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"be80c70f5e8f3ec3160f5ec86288278c73fdc449","unresolved":false,"context_lines":[{"line_number":125,"context_line":"The write performance is not expected to degrade noticeably."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"It is expected any degradation in write performance would be caught by"},{"line_number":128,"context_line":"existing tempest tests."},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"Use of Ilya\u0027s performance tool will be used to verify there is improved read"},{"line_number":131,"context_line":"performance and negligible write performance degradation.[1]"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_04fd4908","line":128,"updated":"2014-08-08 19:57:03.000000000","message":"Tempest seems currently to be a fairly blunt instrument in terms of catching performance degradations, until performance completely falls off a cliff. So I think we\u0027ll probably get a better sense with Ilya\u0027s loadtests.","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"},{"author":{"_account_id":3012,"name":"Dina Belova","email":"dbelova@mirantis.com","username":"dbelova"},"change_message_id":"a71e468266420ba3b3abc473341d65bb4f2b5711","unresolved":false,"context_lines":[{"line_number":127,"context_line":"It is expected any degradation in write performance would be caught by"},{"line_number":128,"context_line":"existing tempest tests."},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"Use of Ilya\u0027s performance tool will be used to verify there is improved read"},{"line_number":131,"context_line":"performance and negligible write performance degradation.[1]"},{"line_number":132,"context_line":""},{"line_number":133,"context_line":"[1] https://github.com/ityaptin/ceilometer/blob/master/tools/sample-generator.py"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_aa8d1bd6","line":130,"updated":"2014-08-11 13:54:01.000000000","message":"Also may you please note Ilya\u0027s IRC nick here as well - just if there will be questions to him after this spec reading not to look for his contacts.","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"be80c70f5e8f3ec3160f5ec86288278c73fdc449","unresolved":false,"context_lines":[{"line_number":130,"context_line":"Use of Ilya\u0027s performance tool will be used to verify there is improved read"},{"line_number":131,"context_line":"performance and negligible write performance degradation.[1]"},{"line_number":132,"context_line":""},{"line_number":133,"context_line":"[1] https://github.com/ityaptin/ceilometer/blob/master/tools/sample-generator.py"},{"line_number":134,"context_line":""},{"line_number":135,"context_line":""},{"line_number":136,"context_line":"Other deployer impact"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_a40d7d37","line":133,"updated":"2014-08-08 19:57:03.000000000","message":"Maybe refer to Ilya\u0027s latest googledoc with instructions?","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"},{"author":{"_account_id":3012,"name":"Dina Belova","email":"dbelova@mirantis.com","username":"dbelova"},"change_message_id":"84c676eb0a73a19c5d64fd41c8717b4427277b16","unresolved":false,"context_lines":[{"line_number":130,"context_line":"Use of Ilya\u0027s performance tool will be used to verify there is improved read"},{"line_number":131,"context_line":"performance and negligible write performance degradation.[1]"},{"line_number":132,"context_line":""},{"line_number":133,"context_line":"[1] https://github.com/ityaptin/ceilometer/blob/master/tools/sample-generator.py"},{"line_number":134,"context_line":""},{"line_number":135,"context_line":""},{"line_number":136,"context_line":"Other deployer impact"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_736142d3","line":133,"in_reply_to":"baada198_a40d7d37","updated":"2014-08-11 11:26:15.000000000","message":"I guess it\u0027ll be good here, yes","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"be80c70f5e8f3ec3160f5ec86288278c73fdc449","unresolved":false,"context_lines":[{"line_number":154,"context_line":"  chungg"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"Other contributors:"},{"line_number":157,"context_line":"  None"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"Ongoing maintainer:"},{"line_number":160,"context_line":"  chungg"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_641fe55a","line":157,"updated":"2014-08-08 19:57:03.000000000","message":"I won\u0027t speak for cdent, but he *might* be interested in having a \"other contributor\" role on this (time/bandwidth allowing etc.)","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"8f38f6163e2d61cdcf9769934d22b8c10bb60252","unresolved":false,"context_lines":[{"line_number":154,"context_line":"  chungg"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"Other contributors:"},{"line_number":157,"context_line":"  None"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"Ongoing maintainer:"},{"line_number":160,"context_line":"  chungg"}],"source_content_type":"text/x-rst","patch_set":10,"id":"baada198_e782f3a3","line":157,"in_reply_to":"baada198_641fe55a","updated":"2014-08-11 10:25:49.000000000","message":"I do seem to be rather drawn to it. My launchpad is chdent. Thanks!","commit_id":"21c0b2f62db7c02c8bb6b850e13c4a4810f06e6e"}]}