)]}'
{"specs/liberty/combine-derived-heterogeneous-meters.rst":[{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"dac1ed0151b8be93c8837a8abdb551793cd3acb5","unresolved":false,"context_lines":[{"line_number":80,"context_line":"                    name: \"cpu.util.memory.usage\""},{"line_number":81,"context_line":"                    unit: \"\""},{"line_number":82,"context_line":"                    type: \"cumulative\""},{"line_number":83,"context_line":"                    expr: \"0.5*$(cpu_util)/100 + 0.5*$(memory.usage)/100\""},{"line_number":84,"context_line":"      publishers:"},{"line_number":85,"context_line":"          - rpc://"},{"line_number":86,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"ba3cc151_765cdf4a","line":83,"updated":"2015-07-03 14:25:19.000000000","message":"how do you ensure the cpu_util and memory usage come from the same resource?\n\nalso, won\u0027t the frequency/timing of the samples affect the validity? how do you ensure the two samples are \u0027insync\u0027 ie. you aren\u0027t building a sample from a fresh memory.usage and old cpu_util.","commit_id":"4ea35346d92b8fc978b22d314e6ebbb976c027be"}]}