)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":30695,"name":"Pedro Henrique Pereira Martins","email":"phpm13@gmail.com","username":"pedrohpmartins"},"change_message_id":"c465c554d878d8345bdbf140ca81bc76b4962f8f","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"0a8ebc67_c3f303b0","updated":"2023-01-27 19:53:40.000000000","message":"Nice change, it seems to be faster using this approach indeed. I have only one question, is there any kind of limitation in the amount of entries we can use in this \u0027in\u0027 clause in the Gnocchi\u0027s search API? is it unlimited? Thanks","commit_id":"80f8a1ca6208ef0922ae41fb2484c7de8665adcb"},{"author":{"_account_id":28356,"name":"Rafael Weingartner","email":"rafael@apache.org","username":"rafaelweingartner"},"change_message_id":"9e345f1a5b1238ee7fcaa8fd96c0fbecefbc9ef7","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"fc489576_36294f78","in_reply_to":"0a8ebc67_c3f303b0","updated":"2023-01-28 17:47:48.000000000","message":"We are shifting the paradigm here. Before this patch, to discover all scopes IDs, one would go through all of Gnocchi resources. However, normally, one will use as the scope attribute the account ID, project ID, or some other identifier that represents the user/owner of resources.\n\nIf we consider a cloud with 1.000.000 different resources registered in Gnocchi (VMs, volumes, IPs, VPN, and so on), the current approach will list all of the 1.000.000 resources to discover all scope IDs. On the other hand, with the proposed solution here, we use the ``search`` method in Gnocchi, and search for all resources where the scope attribute is not in the already found scopes. For the first request, we do not use filter; however, for the subsequent ones, we use the already discovered scopes, and so on. This reduces drastically the number of calls to Gnocchi. Moreover, even for clouds where the number of resources is close to 1.000.000+, the numbers of project IDs, accounts or other identifier will probably be in the 1.000 (probably less), which reduces the number of calls to Gnocchi by 10 folds already.\n\n\nIn the worst case scenario if somebody uses scope attribute as resource ID, then this new approach will be as fast as the previous code. On the other hand, if somebody uses user_id or project_id as the scope attribute, this new method tends to be way faster.","commit_id":"80f8a1ca6208ef0922ae41fb2484c7de8665adcb"},{"author":{"_account_id":15197,"name":"Pierre Riteau","email":"pierre@stackhpc.com","username":"priteau","status":"StackHPC"},"change_message_id":"3846fa2ffc2fed64b9fbd1cf7e82be3fa6c0bdde","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"22640109_af57f5dd","updated":"2023-04-03 13:28:30.000000000","message":"Only changed wrapping / fixed typos.","commit_id":"15be883a1857fc8c10547b95bd72e79e081821df"},{"author":{"_account_id":15197,"name":"Pierre Riteau","email":"pierre@stackhpc.com","username":"priteau","status":"StackHPC"},"change_message_id":"f5179c1d7182b2ca287069b11e8cc3a495e51f70","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"7cb07148_656700ba","updated":"2023-04-03 13:28:25.000000000","message":"Only changed wrapping / fixed typos.","commit_id":"15be883a1857fc8c10547b95bd72e79e081821df"}],"releasenotes/notes/optimize_gnochi-fetcher-runtime-3604026816.yaml":[{"author":{"_account_id":15197,"name":"Pierre Riteau","email":"pierre@stackhpc.com","username":"priteau","status":"StackHPC"},"change_message_id":"e76f96937005a0902183bd5f386ac5018de0a6e4","unresolved":true,"context_lines":[{"line_number":1,"context_line":"---"},{"line_number":2,"context_line":"issues:"},{"line_number":3,"context_line":"  - |"},{"line_number":4,"context_line":"    Optimize Gnocchi fetcher runtime to avoid taking too long when CloudKitty"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"237f1ae4_ff91b078","line":1,"updated":"2023-03-20 14:46:09.000000000","message":"Nit: missing new line at end of file.","commit_id":"80f8a1ca6208ef0922ae41fb2484c7de8665adcb"},{"author":{"_account_id":28356,"name":"Rafael Weingartner","email":"rafael@apache.org","username":"rafaelweingartner"},"change_message_id":"9e33710b29ac14ffa2a99454b99ecc2129ff82ee","unresolved":false,"context_lines":[{"line_number":1,"context_line":"---"},{"line_number":2,"context_line":"issues:"},{"line_number":3,"context_line":"  - |"},{"line_number":4,"context_line":"    Optimize Gnocchi fetcher runtime to avoid taking too long when CloudKitty"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"8233bb0c_78c160cf","line":1,"in_reply_to":"237f1ae4_ff91b078","updated":"2023-03-20 14:51:11.000000000","message":"Done","commit_id":"80f8a1ca6208ef0922ae41fb2484c7de8665adcb"}]}