)]}'
{"specs/victoria/approved/glance_store/spec-lite-ceph-store-optimization.rst":[{"author":{"_account_id":28595,"name":"Victor Coutellier","email":"victor.coutellier@gmail.com","username":"alistarle"},"change_message_id":"49f06cbe5a5e5340d35c066c669ad697b8e99063","unresolved":false,"context_lines":[{"line_number":36,"context_line":"           impact on end users part from improved performance and faster"},{"line_number":37,"context_line":"           operations on their image transfers."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"           Third thing we should finally tackle is the resizes when we do not"},{"line_number":40,"context_line":"           know the image size beforehand. Using just much larger steps when"},{"line_number":41,"context_line":"           resizing the image in Ceph during the write."},{"line_number":42,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"bf51134e_3421e2a0","line":39,"updated":"2020-07-23 19:15:47.000000000","message":"As data always pass trough staging filesystem store, can\u0027t we just read the size of the file in staging in order to create a RBD image of the right size ?","commit_id":"4f6fd2a87243336ea2326142947a7c3e316e19d4"},{"author":{"_account_id":5202,"name":"Erno Kuvaja","email":"jokke@usr.fi","username":"jokke"},"change_message_id":"8b571f74efca68588e0ae8eace74cd5805bf8d8a","unresolved":false,"context_lines":[{"line_number":36,"context_line":"           impact on end users part from improved performance and faster"},{"line_number":37,"context_line":"           operations on their image transfers."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"           Third thing we should finally tackle is the resizes when we do not"},{"line_number":40,"context_line":"           know the image size beforehand. Using just much larger steps when"},{"line_number":41,"context_line":"           resizing the image in Ceph during the write."},{"line_number":42,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_7376e9a6","line":39,"in_reply_to":"bf51134e_3421e2a0","updated":"2020-07-30 14:54:00.000000000","message":"This is not the case always so, no we cannot rely on that. IIUC the case is handled already as in those situations we do know the size of the image.","commit_id":"4f6fd2a87243336ea2326142947a7c3e316e19d4"}],"specs/wallaby/approved/glance_store/spec-lite-ceph-store-optimization.rst":[{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"bf06716464a46fbfd71103653894a23303753761","unresolved":true,"context_lines":[{"line_number":15,"context_line":""},{"line_number":16,"context_line":":solution: Together with sparse images being thinprovisioned by not writing the"},{"line_number":17,"context_line":"           data over the network to the Ceph cluster we probably should utilize"},{"line_number":18,"context_line":"           the possibility that ceph provides of \"buffered write again\" for big"},{"line_number":19,"context_line":"           blops of 00s for those who would prefer the full raw image being"},{"line_number":20,"context_line":"           written to the disk to ease the network utilization. This specially"},{"line_number":21,"context_line":"           becomes meaningful together with multi-store and edge where the"}],"source_content_type":"text/x-rst","patch_set":6,"id":"58678e41_8f4ae718","line":18,"range":{"start_line":18,"start_character":49,"end_line":18,"end_character":71},"updated":"2021-01-21 14:09:49.000000000","message":"I\u0027d suggest saying something about the support of this feature (that is, does it require a fairly new Ceph installation, or has it been around for a long time?).","commit_id":"d95147355d184c045dd08a03434286b021c4dc82"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"7439615eced312d0eaff125a1fe934cf774ae6bd","unresolved":true,"context_lines":[{"line_number":16,"context_line":":solution: Together with sparse images being thinprovisioned by not writing the"},{"line_number":17,"context_line":"           data over the network to the Ceph cluster we probably should utilize"},{"line_number":18,"context_line":"           the possibility that ceph provides of \"buffered write again\" for big"},{"line_number":19,"context_line":"           blops of 00s for those who would prefer the full raw image being"},{"line_number":20,"context_line":"           written to the disk to ease the network utilization. This specially"},{"line_number":21,"context_line":"           becomes meaningful together with multi-store and edge where the"},{"line_number":22,"context_line":"           store we are writing to might be behind high latency link. Even the"}],"source_content_type":"text/x-rst","patch_set":6,"id":"4ea92064_0b82250b","line":19,"range":{"start_line":19,"start_character":11,"end_line":19,"end_character":16},"updated":"2021-01-28 14:55:02.000000000","message":"blobs","commit_id":"d95147355d184c045dd08a03434286b021c4dc82"},{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"bf06716464a46fbfd71103653894a23303753761","unresolved":true,"context_lines":[{"line_number":21,"context_line":"           becomes meaningful together with multi-store and edge where the"},{"line_number":22,"context_line":"           store we are writing to might be behind high latency link. Even the"},{"line_number":23,"context_line":"           multi-store operations are mostly under the async nature of taskflow"},{"line_number":24,"context_line":"           speeding things up and not consuming the network capasity from other"},{"line_number":25,"context_line":"           operations would be beneficial."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"           This approache should not danger data integrity nor have an impact on"}],"source_content_type":"text/x-rst","patch_set":6,"id":"b9556316_e0b20c9c","line":24,"range":{"start_line":24,"start_character":60,"end_line":24,"end_character":68},"updated":"2021-01-21 14:09:49.000000000","message":"capacity","commit_id":"d95147355d184c045dd08a03434286b021c4dc82"},{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"bf06716464a46fbfd71103653894a23303753761","unresolved":true,"context_lines":[{"line_number":24,"context_line":"           speeding things up and not consuming the network capasity from other"},{"line_number":25,"context_line":"           operations would be beneficial."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"           This approache should not danger data integrity nor have an impact on"},{"line_number":28,"context_line":"           end users\u0027 part from improved performance and faster operations on"},{"line_number":29,"context_line":"           their image transfers."},{"line_number":30,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"bc5bfac0_91903b86","line":27,"range":{"start_line":27,"start_character":16,"end_line":27,"end_character":25},"updated":"2021-01-21 14:09:49.000000000","message":"approach","commit_id":"d95147355d184c045dd08a03434286b021c4dc82"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"7439615eced312d0eaff125a1fe934cf774ae6bd","unresolved":true,"context_lines":[{"line_number":24,"context_line":"           speeding things up and not consuming the network capasity from other"},{"line_number":25,"context_line":"           operations would be beneficial."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"           This approache should not danger data integrity nor have an impact on"},{"line_number":28,"context_line":"           end users\u0027 part from improved performance and faster operations on"},{"line_number":29,"context_line":"           their image transfers."},{"line_number":30,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"746fd193_d3b90d54","line":27,"range":{"start_line":27,"start_character":37,"end_line":27,"end_character":43},"updated":"2021-01-28 14:55:02.000000000","message":"endanger","commit_id":"d95147355d184c045dd08a03434286b021c4dc82"},{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"bf06716464a46fbfd71103653894a23303753761","unresolved":true,"context_lines":[{"line_number":25,"context_line":"           operations would be beneficial."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"           This approache should not danger data integrity nor have an impact on"},{"line_number":28,"context_line":"           end users\u0027 part from improved performance and faster operations on"},{"line_number":29,"context_line":"           their image transfers."},{"line_number":30,"context_line":""},{"line_number":31,"context_line":":impacts: None"}],"source_content_type":"text/x-rst","patch_set":6,"id":"d6b22ab7_8762dfb4","line":28,"range":{"start_line":28,"start_character":11,"end_line":28,"end_character":31},"updated":"2021-01-21 14:09:49.000000000","message":"maybe: end users other than","commit_id":"d95147355d184c045dd08a03434286b021c4dc82"},{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"bf06716464a46fbfd71103653894a23303753761","unresolved":true,"context_lines":[{"line_number":35,"context_line":""},{"line_number":36,"context_line":"               Problem with the async writes is that we would need to implement"},{"line_number":37,"context_line":"               chunk pool tracing ourselves and it does not address the wasteful"},{"line_number":38,"context_line":"               badwidth usage, just potentially utilizes the available network"},{"line_number":39,"context_line":"               more aggressively."},{"line_number":40,"context_line":""},{"line_number":41,"context_line":":timeline: Target on milestone 3"}],"source_content_type":"text/x-rst","patch_set":6,"id":"28a4b1d1_e26e8819","line":38,"range":{"start_line":38,"start_character":15,"end_line":38,"end_character":23},"updated":"2021-01-21 14:09:49.000000000","message":"bandwidth","commit_id":"d95147355d184c045dd08a03434286b021c4dc82"}]}