)]}'
{"specs/liberty/distributed-coordinated-notifications.rst":[{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"cb26b470dfaa163e7f140eef2b72407d2a343188","unresolved":false,"context_lines":[{"line_number":31,"context_line":"Data can be handled in parallel even at the pipeline level. For example, when"},{"line_number":32,"context_line":"there are no transformers, datapoints do not need to be handled sequentially."},{"line_number":33,"context_line":"Additionally, when transformers are present, datapoints of different resources"},{"line_number":34,"context_line":"have no relevance to each other and can be handled in parallel."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"Proposed change"},{"line_number":37,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":1,"id":"ba3cc151_98b250e4","line":34,"updated":"2015-07-08 13:37:07.000000000","message":"I would guess we need to confirm that this is in fact true. Is it possible to have a transformer that aggregates some values across project or user id?\n\nIt\u0027s conceivable that someone could cook that, but it seems pretty high to do it in the pipeline rather than in post processing.","commit_id":"5c03c013c7eb059e10b9512eab98715359286e9d"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"54f745ca0c5d588421c05f76be3f803e2a2d3250","unresolved":false,"context_lines":[{"line_number":31,"context_line":"Data can be handled in parallel even at the pipeline level. For example, when"},{"line_number":32,"context_line":"there are no transformers, datapoints do not need to be handled sequentially."},{"line_number":33,"context_line":"Additionally, when transformers are present, datapoints of different resources"},{"line_number":34,"context_line":"have no relevance to each other and can be handled in parallel."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"Proposed change"},{"line_number":37,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":1,"id":"ba3cc151_ac0ef73a","line":34,"in_reply_to":"ba3cc151_98b250e4","updated":"2015-07-08 14:13:15.000000000","message":"alternatively we can make this configurable (one or more of: resource/user/project). but that requires the user to understand what it actually means.","commit_id":"5c03c013c7eb059e10b9512eab98715359286e9d"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"cb26b470dfaa163e7f140eef2b72407d2a343188","unresolved":false,"context_lines":[{"line_number":44,"context_line":"The basic workflow is as follows:"},{"line_number":45,"context_line":"- on notification agent startup, create a listener for main queue"},{"line_number":46,"context_line":"- for each pipeline definition, we create x queues and x listeners where x"},{"line_number":47,"context_line":"  corresponds to the number of notification agents registered to group"},{"line_number":48,"context_line":"- when a datapoint is received, agent builds sample."},{"line_number":49,"context_line":"- after sample is built, we hash (or don\u0027t hash) resource id (mandatory)"},{"line_number":50,"context_line":"  and mod by number of agents."}],"source_content_type":"text/x-rst","patch_set":1,"id":"ba3cc151_b3a2d5ac","line":47,"updated":"2015-07-08 13:37:07.000000000","message":"Given that we\u0027ll need to update queues and listeners as agents leave and join the coordination mechanism which will leave some transformations in an unresolved state. We\u0027ve already got this now, so I\u0027m not sure it is a big deal, but might be useful to at least ack it.","commit_id":"5c03c013c7eb059e10b9512eab98715359286e9d"}]}