Add perf metrics for 2.23.0

[khluu]

REGRESSION 12.60%: client__tasks_and_get_batch (THROUGHPUT) regresses from 0.9229179501706766 to 0.8066078636677214 in microbenchmark.json
REGRESSION 9.95%: single_client_tasks_async (THROUGHPUT) regresses from 8194.32696391844 to 7378.7495889110505 in microbenchmark.json
REGRESSION 7.64%: multi_client_put_calls_Plasma_Store (THROUGHPUT) regresses from 12873.027401685875 to 11889.62967857741 in microbenchmark.json
REGRESSION 6.02%: placement_group_create/removal (THROUGHPUT) regresses from 838.5131869401664 to 788.0742341227435 in microbenchmark.json
REGRESSION 5.08%: single_client_tasks_and_get_batch (THROUGHPUT) regresses from 8.144156516407913 to 7.730134993824188 in microbenchmark.json
REGRESSION 4.34%: 1_1_actor_calls_sync (THROUGHPUT) regresses from 2096.4576697416196 to 2005.483489295091 in microbenchmark.json
REGRESSION 4.00%: client__tasks_and_put_batch (THROUGHPUT) regresses from 11657.997816451738 to 11191.448976686781 in microbenchmark.json
REGRESSION 3.82%: actors_per_second (THROUGHPUT) regresses from 638.1892393086097 to 613.8104422632972 in benchmarks/many_actors.json
REGRESSION 3.27%: 1_1_actor_calls_async (THROUGHPUT) regresses from 9062.792852859755 to 8766.429278436297 in microbenchmark.json
REGRESSION 3.14%: client__get_calls (THROUGHPUT) regresses from 1158.7613618440014 to 1122.3334401198258 in microbenchmark.json
REGRESSION 2.21%: 1_n_async_actor_calls_async (THROUGHPUT) regresses from 7460.166048300169 to 7295.2280994398525 in microbenchmark.json
REGRESSION 1.52%: n_n_actor_calls_with_arg_async (THROUGHPUT) regresses from 2713.504279822532 to 2672.287170984248 in microbenchmark.json
REGRESSION 1.34%: 1_n_actor_calls_async (THROUGHPUT) regresses from 8606.08018065877 to 8490.352729000244 in microbenchmark.json
REGRESSION 1.32%: n_n_actor_calls_async (THROUGHPUT) regresses from 27688.27595653398 to 27321.88100222539 in microbenchmark.json
REGRESSION 0.62%: pgs_per_second (THROUGHPUT) regresses from 23.57570074520289 to 23.42930283459158 in benchmarks/many_pgs.json
REGRESSION 0.29%: client__put_gigabytes (THROUGHPUT) regresses from 0.13155858254345548 to 0.13117998295265082 in microbenchmark.json
REGRESSION 0.25%: 1_1_actor_calls_concurrent (THROUGHPUT) regresses from 5479.71996068499 to 5466.2521928780425 in microbenchmark.json
REGRESSION 0.15%: tasks_per_second (THROUGHPUT) regresses from 580.7153723158294 to 579.8482552816317 in benchmarks/many_tasks.json
REGRESSION 35.76%: stage_2_avg_iteration_time (LATENCY) regresses from 54.19231338500977 to 73.57129535675048 in stress_tests/stress_test_many_tasks.json
REGRESSION 25.91%: dashboard_p50_latency_ms (LATENCY) regresses from 68.347 to 86.057 in benchmarks/many_actors.json
REGRESSION 16.53%: stage_1_avg_iteration_time (LATENCY) regresses from 23.831846237182617 to 27.771051788330077 in stress_tests/stress_test_many_tasks.json
REGRESSION 16.02%: dashboard_p95_latency_ms (LATENCY) regresses from 2276.742 to 2641.416 in benchmarks/many_actors.json
REGRESSION 13.10%: dashboard_p99_latency_ms (LATENCY) regresses from 4468.284 to 5053.446 in benchmarks/many_actors.json
REGRESSION 12.51%: stage_3_time (LATENCY) regresses from 2976.5505974292755 to 3348.829131603241 in stress_tests/stress_test_many_tasks.json
REGRESSION 10.60%: dashboard_p99_latency_ms (LATENCY) regresses from 812.483 to 898.617 in benchmarks/many_pgs.json
REGRESSION 4.98%: avg_pg_create_time_ms (LATENCY) regresses from 0.8636297822809175 to 0.9066000405417517 in stress_tests/stress_test_placement_group.json
REGRESSION 3.78%: stage_0_time (LATENCY) regresses from 12.160747766494751 to 12.619915246963501 in stress_tests/stress_test_many_tasks.json
REGRESSION 3.72%: dashboard_p50_latency_ms (LATENCY) regresses from 42.738 to 44.327 in benchmarks/many_tasks.json
REGRESSION 3.25%: 10000_get_time (LATENCY) regresses from 23.237287506 to 23.992132071 in scalability/single_node.json
REGRESSION 1.79%: dashboard_p95_latency_ms (LATENCY) regresses from 3207.541 to 3265.071 in benchmarks/many_tasks.json
REGRESSION 0.85%: 10000_args_time (LATENCY) regresses from 17.131764829000005 to 17.276734743999995 in scalability/single_node.json
REGRESSION 0.76%: 107374182400_large_object_time (LATENCY) regresses from 31.629831378000006 to 31.869567716000006 in scalability/single_node.json

[jjyao]

LG