[
  {
    "50.00 percentile latency (ns)": 47071283648,
    "90.00 percentile latency (ns)": 68694863629,
    "95.00 percentile latency (ns)": 70405650606,
    "97.00 percentile latency (ns)": 71116535607,
    "99.00 percentile latency (ns)": 71691388352,
    "99.90 percentile latency (ns)": 71786718527,
    "Max latency (ns)": 71786718527,
    "Mean latency (ns)": 43544191111,
    "Min duration satisfied": "Yes",
    "Min latency (ns)": 1067464518,
    "Min queries satisfied": "Yes",
    "Mode": "Performance",
    "Result is": "VALID",
    "SUT name": "GNMT SERVER",
    "Samples per second": 342.347,
    "Scenario": "offline",
    "accelerator_memory_capacity": "16GB",
    "accelerator_memory_configuration": "GDDR6",
    "accelerator_model_name": "Nvidia Tesla T4",
    "accelerators_per_node": 1,
    "accuracy_log_probability": 0,
    "accuracy_log_rng_seed": 0,
    "characteristics.blue": 23.8,
    "characteristics.samples_per_second": 342.347,
    "characteristics.samples_per_second.normalized_per_core": 342.347,
    "characteristics.samples_per_second.normalized_per_processor": 342.347,
    "ck_system": "alibaba_cloud_t4",
    "ck_used": true,
    "dataset": "WMT E-G",
    "dataset_link": "",
    "dim_x_default": "characteristics.samples_per_second",
    "dim_x_maximize": true,
    "dim_y_default": "characteristics.blue",
    "dim_y_maximize": true,
    "division": "closed",
    "formal_model": "nmt",
    "formal_model_accuracy": 99.0,
    "formal_model_link": "",
    "framework": "TensorRT 6.0, CUDA 10.1, cuDNN 7.6.3, libjemalloc2, cub 1.8.0, tensorrt-laboratory mlperf branch",
    "host_memory_capacity": "93GB",
    "host_processor_core_count": 120,
    "host_processor_model_name": "AMD EPYC 7V13 64-Core Processor",
    "host_processor_vcpu_count": 24,
    "host_processors_per_node": 1,
    "host_storage_capacity": "500GB",
    "host_storage_type": "SSD",
    "informal_model": "gnmt",
    "input_data_types": "UTF-8",
    "key.accuracy": "characteristics.blue",
    "max_async_queries": 1,
    "max_duration (ms)": 0,
    "max_query_count": 0,
    "min_duration (ms)": 60000,
    "min_query_count": 1,
    "mlperf_version": 0.5,
    "normalize_cores": 1,
    "normalize_processors": 1,
    "note_code": "https://github.com/mlcommons/inference_results_v0.5/tree/master/closed/Alibaba/code",
    "note_details": "https://github.com/mlcommons/inference_results_v0.5/tree/master/closed/Alibaba/results/alibaba_cloud_t4",
    "number_of_nodes": 1,
    "operating_system": "Ubuntu 18.04.5 LTS (Linux-5.4.0-1055-azure-x86_64-with-Ubuntu-18.04-bionic)",
    "other_software_stack": "docker 18.09.2, python 3.6.8,gcc 5.5.0,onnx 1.3.0, tensorflow 1.13.1, pytorch 1.1.0, torchvision 0.3.0, pycuda 2019.1, sacrebleu 1.3.3, SimpleJSON, OpenCV 4.1.1; GCC 7.5.0; Python 3.7.10",
    "performance_issue_same": true,
    "performance_issue_same_index": 0,
    "performance_issue_unique": true,
    "performance_sample_count": 3903900,
    "print_timestamps": true,
    "problem": false,
    "qsl_rng_seed": 3133965575612453542,
    "retraining": "N",
    "sample_index_rng_seed": 665484352860916858,
    "samples_per_query": 24576,
    "schedule_rng_seed": 3622009729038561421,
    "starting_weights_filename": "ende_gnmt_model_4_layer/translate.ckpt.data-00000-of-00001",
    "status": "available",
    "submitter": "Alibaba",
    "submitter_link": "https://github.com/ctuning/ck-mlperf-inference/tree/main/bench.mlperf.submitter/Alibaba",
    "sw_notes": "Powered by CK v2.5.8 (https://github.com/ctuning/ck)",
    "system_link": "https://github.com/ctuning/ck-mlperf-inference/tree/main/bench.mlperf.system/alibaba_cloud_t4",
    "system_name": "Microsoft Corporation 7.0 (Virtual Machine)",
    "target_latency (ns)": 0,
    "target_qps": 340,
    "task": "NMT",
    "task2": "nmt",
    "total_cores": 120,
    "uid": "a583b1a4b4882b07",
    "use_accelerator": true,
    "weight_data_types": "int8,fp16",
    "weight_transformations": "quantization, affine fusion"
  }
]