)]}'
{"doc/source/measurements.rst":[{"author":{"_account_id":2860,"name":"litong01","email":"email4tong@yahoo.com","username":"litong01"},"change_message_id":"51141a5e347f061a21365cc1a73e38c5068ac7ea","unresolved":false,"context_lines":[{"line_number":71,"context_line":"instance                       Gauge       instance   inst ID   both          Existence of instance"},{"line_number":72,"context_line":"instance:\u003ctype\u003e                Gauge       instance   inst ID   both          Existence of instance \u003ctype\u003e (openstack types)"},{"line_number":73,"context_line":"memory                         Gauge       MB         inst ID   notification  Volume of RAM in MB"},{"line_number":74,"context_line":"cpu                            Cumulative  ns         inst ID   pollster      The real CPU time used by the instance (uptime * CPU utilisation)"},{"line_number":75,"context_line":"cpu_util                       Gauge       %          inst ID   pollster      Average CPU utilisation"},{"line_number":76,"context_line":"vcpus                          Gauge       vcpu       inst ID   notification  Number of VCPUs"},{"line_number":77,"context_line":"disk.read.requests             Cumulative  request    inst ID   pollster      Number of read requests"}],"source_content_type":"text/x-rst","patch_set":2,"id":"AAAAUX%2F%2F0yI%3D","line":74,"updated":"2014-02-10 15:41:14.000000000","message":"Not sure about the word \"real\", is it really necessary? it will be needed if there\nis a not-real case, but I do not see where we track these. would it be better\nto have the followings:\n\n    1. The CPU time used by the instance ...\n or \n    2. The actual CPU time used by the instance ...\n\nThanks.","commit_id":"d85d8ed30d72d6b1f19c67d65d539aa4967ab4dd"},{"author":{"_account_id":1669,"name":"Julien Danjou","display_name":"jd","email":"julien@danjou.info","username":"jdanjou"},"change_message_id":"baf6f5046b35de6d0c5f4a3f79ee0e9f0353ae2f","unresolved":false,"context_lines":[{"line_number":71,"context_line":"instance                       Gauge       instance   inst ID   both          Existence of instance"},{"line_number":72,"context_line":"instance:\u003ctype\u003e                Gauge       instance   inst ID   both          Existence of instance \u003ctype\u003e (openstack types)"},{"line_number":73,"context_line":"memory                         Gauge       MB         inst ID   notification  Volume of RAM in MB"},{"line_number":74,"context_line":"cpu                            Cumulative  ns         inst ID   pollster      The CPU time used by the instance (uptime * CPU utilisation)"},{"line_number":75,"context_line":"cpu_util                       Gauge       %          inst ID   pollster      Average CPU utilisation"},{"line_number":76,"context_line":"vcpus                          Gauge       vcpu       inst ID   notification  Number of VCPUs"},{"line_number":77,"context_line":"disk.read.requests             Cumulative  request    inst ID   pollster      Number of read requests"}],"source_content_type":"text/x-rst","patch_set":3,"id":"AAAAUX%2F%2F0PQ%3D","line":74,"updated":"2014-02-10 16:39:32.000000000","message":"I don\u0027t think it improves anything?","commit_id":"4778bd4bd7b9dcff05ec1c0387fa2a6df55f5761"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"5e69fe5d804d8db3a906088b451048f28747b660","unresolved":false,"context_lines":[{"line_number":71,"context_line":"instance                       Gauge       instance   inst ID   both          Existence of instance"},{"line_number":72,"context_line":"instance:\u003ctype\u003e                Gauge       instance   inst ID   both          Existence of instance \u003ctype\u003e (openstack types)"},{"line_number":73,"context_line":"memory                         Gauge       MB         inst ID   notification  Volume of RAM in MB"},{"line_number":74,"context_line":"cpu                            Cumulative  ns         inst ID   pollster      The CPU time used by the instance (uptime * CPU utilisation)"},{"line_number":75,"context_line":"cpu_util                       Gauge       %          inst ID   pollster      Average CPU utilisation"},{"line_number":76,"context_line":"vcpus                          Gauge       vcpu       inst ID   notification  Number of VCPUs"},{"line_number":77,"context_line":"disk.read.requests             Cumulative  request    inst ID   pollster      Number of read requests"}],"source_content_type":"text/x-rst","patch_set":3,"id":"AAAAUX%2F%2FuAE%3D","line":74,"in_reply_to":"AAAAUX%2F%2F0PQ%3D","updated":"2014-02-11 08:51:51.000000000","message":"I think the uptime * CPU utilisation gives extra information. \n\nThe idea to enhance the documentation is coming from an irc discussion http://eavesdrop.openstack.org/irclogs/%23openstack-ceilometer/%23openstack-ceilometer.2014-01-28.log at 2014-01-28T15:33:12\n\nI\u0027m open to suggestion how to enhance the information in this documentation.","commit_id":"4778bd4bd7b9dcff05ec1c0387fa2a6df55f5761"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"11fdaa7b47f433a95c7f70d69c155d916c373e74","unresolved":false,"context_lines":[{"line_number":71,"context_line":"instance                       Gauge       instance   inst ID   both          Existence of instance"},{"line_number":72,"context_line":"instance:\u003ctype\u003e                Gauge       instance   inst ID   both          Existence of instance \u003ctype\u003e (openstack types)"},{"line_number":73,"context_line":"memory                         Gauge       MB         inst ID   notification  Volume of RAM in MB"},{"line_number":74,"context_line":"cpu                            Cumulative  ns         inst ID   pollster      The CPU time used by the instance (uptime * CPU utilisation)"},{"line_number":75,"context_line":"cpu_util                       Gauge       %          inst ID   pollster      Average CPU utilisation"},{"line_number":76,"context_line":"vcpus                          Gauge       vcpu       inst ID   notification  Number of VCPUs"},{"line_number":77,"context_line":"disk.read.requests             Cumulative  request    inst ID   pollster      Number of read requests"}],"source_content_type":"text/x-rst","patch_set":3,"id":"AAAAUX%2F%2Fa9M%3D","line":74,"in_reply_to":"AAAAUX%2F%2Fr3I%3D","updated":"2014-02-13 11:28:46.000000000","message":"Yes, my understanding is similar so that the cpu_util meter is scaled by the number of vcpus and the load on the vcpus. If I understand my measurement results correctly then the cpu meter depends on all those three factors (wall clock uptime, number of vcpus, load on those vcpus).\n\nI will try to capture this information in the next patchset.","commit_id":"4778bd4bd7b9dcff05ec1c0387fa2a6df55f5761"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"cc5e0ee2090a73b98629298b9c393d30c8b500d8","unresolved":false,"context_lines":[{"line_number":71,"context_line":"instance                       Gauge       instance   inst ID   both          Existence of instance"},{"line_number":72,"context_line":"instance:\u003ctype\u003e                Gauge       instance   inst ID   both          Existence of instance \u003ctype\u003e (openstack types)"},{"line_number":73,"context_line":"memory                         Gauge       MB         inst ID   notification  Volume of RAM in MB"},{"line_number":74,"context_line":"cpu                            Cumulative  ns         inst ID   pollster      The CPU time used by the instance (uptime * CPU utilisation)"},{"line_number":75,"context_line":"cpu_util                       Gauge       %          inst ID   pollster      Average CPU utilisation"},{"line_number":76,"context_line":"vcpus                          Gauge       vcpu       inst ID   notification  Number of VCPUs"},{"line_number":77,"context_line":"disk.read.requests             Cumulative  request    inst ID   pollster      Number of read requests"}],"source_content_type":"text/x-rst","patch_set":3,"id":"AAAAUX%2F%2Fr3I%3D","line":74,"in_reply_to":"AAAAUX%2F%2FuAE%3D","updated":"2014-02-11 14:23:20.000000000","message":"Note that CPU utilization (in the sense of the cpu_util meter) is scaled by the number of vcpus assigned to the instance, whereas uptime is simply cumulative wall clock time IIUC.","commit_id":"4778bd4bd7b9dcff05ec1c0387fa2a6df55f5761"}]}