perf_analyzer
No description available.
File Explorer
Download Latest Version (.zip)- devcontainer.json
- codeql.yml
- mirror_repo.sh
- pre-commit.yml
- python-package-genai.yml
- trigger_ci.yml
- benchmarking.md
- cli.md
- inference_load_modes.md
- input_data.md
- install.md
- measurements_metrics.md
- quick_start.md
- README.md
- custom_gpu_metrics.csv
- distribution_of_input_sequence_lengths_to_output_sequence_lengths.jpeg
- request_latency.jpeg
- time_to_first_token.jpeg
- time_to_first_token_vs_input_sequence_lengths.jpeg
- token-to-token_latency_vs_output_token_position.jpeg
- analyze.md
- customizable_frontends.md
- customizable_payloads.md
- dynamo.md
- embeddings.md
- example_plots.md
- files.md
- goodput.md
- gpu_telemetry.md
- huggingface_tgi.md
- lora.md
- multi_modal.md
- multi_turn.md
- process_export_files.md
- rankings.md
- tutorial.md
- checkpoint.py
- genai_perf_config.py
- objective_parameter.py
- perf_analyzer_config.py
- search_parameter.py
- search_parameters.py
- sweep_objective_generator.py
- base_config.py
- config_analyze.py
- config_command.py
- config_defaults.py
- config_endpoint.py
- config_field.py
- config_input.py
- config_output.py
- config_perf_analyzer.py
- config_process.py
- config_tokenizer.py
- create_config.py
- results.py
- run_config.py
- endpoint_config.py
- model_spec.py
- console_exporter.py
- csv_exporter.py
- data_exporter_factory.py
- data_exporter_interface.py
- exporter_config.py
- exporter_utils.py
- json_exporter.py
- output_reporter.py
- telemetry_data_exporter_util.py
- __init__.py
- goodput_calculator.py
- llm_goodput_calculator.py
- __init__.py
- base_converter.py
- dynamic_grpc_converter.py
- huggingface_generate_converter.py
- image_retrieval_converter.py
- nvclip_converter.py
- openai_chat_completions_converter.py
- openai_completions_converter.py
- openai_embeddings_converter.py
- output_format_converter_factory.py
- rankings_converter.py
- template_converter.py
- tensorrtllm_converter.py
- tensorrtllm_engine_converter.py
- triton_generate_converter.py
- vllm_converter.py
- dlss.png
- h100.jpeg
- h200.jpeg
- jensen.jpeg
- __init__.py
- base_file_input_retriever.py
- base_input_retriever.py
- file_input_retriever.py
- generic_dataset.py
- input_retriever_factory.py
- payload_input_retriever.py
- shakespeare.txt
- synthetic_audio_generator.py
- synthetic_data_retriever.py
- synthetic_image_generator.py
- synthetic_prompt_generator.py
- __init__.py
- input_constants.py
- inputs.py
- inputs_config.py
- model_config_measurement.py
- model_constraints.py
- run_config_measurement.py
- run_constraints.py
- __init__.py
- image_retrieval_metrics.py
- llm_metrics.py
- metrics.py
- statistics.py
- telemetry_metrics.py
- telemetry_statistics.py
- telemetry_stats_aggregator.py
- __init__.py
- base_plot.py
- box_plot.py
- heat_map.py
- plot_config.py
- plot_config_parser.py
- plot_manager.py
- scatter_plot.py
- __init__.py
- image_retrieval_profile_data_parser.py
- llm_profile_data_parser.py
- merged_profile_parser.py
- parser_result.py
- profile_data_parser.py
- __init__.py
- energy_consumption_avg.py
- energy_consumption_base.py
- energy_consumption_max.py
- energy_consumption_min.py
- energy_consumption_p1.py
- energy_consumption_p10.py
- energy_consumption_p25.py
- energy_consumption_p5.py
- energy_consumption_p50.py
- energy_consumption_p75.py
- energy_consumption_p90.py
- energy_consumption_p95.py
- energy_consumption_p99.py
- energy_consumption_std.py
- gpu_clock_memory_avg.py
- gpu_clock_memory_base.py
- gpu_clock_memory_max.py
- gpu_clock_memory_min.py
- gpu_clock_memory_p1.py
- gpu_clock_memory_p10.py
- gpu_clock_memory_p25.py
- gpu_clock_memory_p5.py
- gpu_clock_memory_p50.py
- gpu_clock_memory_p75.py
- gpu_clock_memory_p90.py
- gpu_clock_memory_p95.py
- gpu_clock_memory_p99.py
- gpu_clock_memory_std.py
- gpu_clock_sm_avg.py
- gpu_clock_sm_base.py
- gpu_clock_sm_max.py
- gpu_clock_sm_min.py
- gpu_clock_sm_p1.py
- gpu_clock_sm_p10.py
- gpu_clock_sm_p25.py
- gpu_clock_sm_p5.py
- gpu_clock_sm_p50.py
- gpu_clock_sm_p75.py
- gpu_clock_sm_p90.py
- gpu_clock_sm_p95.py
- gpu_clock_sm_p99.py
- gpu_clock_sm_std.py
- gpu_energy_consumption_avg.py
- gpu_memory_free_avg.py
- gpu_memory_free_base.py
- gpu_memory_free_max.py
- gpu_memory_free_min.py
- gpu_memory_free_p1.py
- gpu_memory_free_p10.py
- gpu_memory_free_p25.py
- gpu_memory_free_p5.py
- gpu_memory_free_p50.py
- gpu_memory_free_p75.py
- gpu_memory_free_p90.py
- gpu_memory_free_p95.py
- gpu_memory_free_p99.py
- gpu_memory_free_std.py
- gpu_memory_temperature_avg.py
- gpu_memory_temperature_base.py
- gpu_memory_temperature_max.py
- gpu_memory_temperature_min.py
- gpu_memory_temperature_p1.py
- gpu_memory_temperature_p10.py
- gpu_memory_temperature_p25.py
- gpu_memory_temperature_p5.py
- gpu_memory_temperature_p50.py
- gpu_memory_temperature_p75.py
- gpu_memory_temperature_p90.py
- gpu_memory_temperature_p95.py
- gpu_memory_temperature_p99.py
- gpu_memory_temperature_std.py
- gpu_memory_used_avg.py
- gpu_memory_used_base.py
- gpu_memory_used_max.py
- gpu_memory_used_min.py
- gpu_memory_used_p1.py
- gpu_memory_used_p10.py
- gpu_memory_used_p25.py
- gpu_memory_used_p5.py
- gpu_memory_used_p50.py
- gpu_memory_used_p75.py
- gpu_memory_used_p90.py
- gpu_memory_used_p95.py
- gpu_memory_used_p99.py
- gpu_memory_used_std.py
- gpu_power_limit_avg.py
- gpu_power_limit_base.py
- gpu_power_usage_avg.py
- gpu_power_usage_base.py
- gpu_power_usage_max.py
- gpu_power_usage_min.py
- gpu_power_usage_p1.py
- gpu_power_usage_p10.py
- gpu_power_usage_p25.py
- gpu_power_usage_p5.py
- gpu_power_usage_p50.py
- gpu_power_usage_p75.py
- gpu_power_usage_p90.py
- gpu_power_usage_p95.py
- gpu_power_usage_p99.py
- gpu_power_usage_std.py
- gpu_temperature_avg.py
- gpu_temperature_base.py
- gpu_temperature_max.py
- gpu_temperature_min.py
- gpu_temperature_p1.py
- gpu_temperature_p10.py
- gpu_temperature_p25.py
- gpu_temperature_p5.py
- gpu_temperature_p50.py
- gpu_temperature_p75.py
- gpu_temperature_p90.py
- gpu_temperature_p95.py
- gpu_temperature_p99.py
- gpu_temperature_std.py
- gpu_utilization_avg.py
- gpu_utilization_base.py
- gpu_utilization_max.py
- gpu_utilization_min.py
- gpu_utilization_p1.py
- gpu_utilization_p10.py
- gpu_utilization_p25.py
- gpu_utilization_p5.py
- gpu_utilization_p50.py
- gpu_utilization_p75.py
- gpu_utilization_p90.py
- gpu_utilization_p95.py
- gpu_utilization_p99.py
- gpu_utilization_std.py
- input_sequence_length_avg.py
- input_sequence_length_base.py
- input_sequence_length_max.py
- input_sequence_length_min.py
- input_sequence_length_p1.py
- input_sequence_length_p10.py
- input_sequence_length_p25.py
- input_sequence_length_p5.py
- input_sequence_length_p50.py
- input_sequence_length_p75.py
- input_sequence_length_p90.py
- input_sequence_length_p95.py
- input_sequence_length_p99.py
- input_sequence_length_std.py
- inter_token_latency_avg.py
- inter_token_latency_base.py
- inter_token_latency_max.py
- inter_token_latency_min.py
- inter_token_latency_p1.py
- inter_token_latency_p10.py
- inter_token_latency_p25.py
- inter_token_latency_p5.py
- inter_token_latency_p50.py
- inter_token_latency_p75.py
- inter_token_latency_p90.py
- inter_token_latency_p95.py
- inter_token_latency_p99.py
- inter_token_latency_std.py
- memory_copy_utilization_avg.py
- memory_copy_utilization_base.py
- memory_copy_utilization_max.py
- memory_copy_utilization_min.py
- memory_copy_utilization_p1.py
- memory_copy_utilization_p10.py
- memory_copy_utilization_p25.py
- memory_copy_utilization_p5.py
- memory_copy_utilization_p50.py
- memory_copy_utilization_p75.py
- memory_copy_utilization_p90.py
- memory_copy_utilization_p95.py
- memory_copy_utilization_p99.py
- memory_copy_utilization_std.py
- output_sequence_length_avg.py
- output_sequence_length_base.py
- output_sequence_length_max.py
- output_sequence_length_min.py
- output_sequence_length_p1.py
- output_sequence_length_p10.py
- output_sequence_length_p25.py
- output_sequence_length_p5.py
- output_sequence_length_p50.py
- output_sequence_length_p75.py
- output_sequence_length_p90.py
- output_sequence_length_p95.py
- output_sequence_length_p99.py
- output_sequence_length_std.py
- output_token_throughput_avg.py
- output_token_throughput_per_user_avg.py
- output_token_throughput_per_user_base.py
- output_token_throughput_per_user_max.py
- output_token_throughput_per_user_min.py
- output_token_throughput_per_user_p1.py
- output_token_throughput_per_user_p10.py
- output_token_throughput_per_user_p25.py
- output_token_throughput_per_user_p5.py
- output_token_throughput_per_user_p50.py
- output_token_throughput_per_user_p75.py
- output_token_throughput_per_user_p90.py
- output_token_throughput_per_user_p95.py
- output_token_throughput_per_user_p99.py
- output_token_throughput_per_user_std.py
- pcie_receive_throughput_avg.py
- pcie_receive_throughput_base.py
- pcie_receive_throughput_max.py
- pcie_receive_throughput_min.py
- pcie_receive_throughput_p1.py
- pcie_receive_throughput_p10.py
- pcie_receive_throughput_p25.py
- pcie_receive_throughput_p5.py
- pcie_receive_throughput_p50.py
- pcie_receive_throughput_p75.py
- pcie_receive_throughput_p90.py
- pcie_receive_throughput_p95.py
- pcie_receive_throughput_p99.py
- pcie_receive_throughput_std.py
- pcie_replay_counter_avg.py
- pcie_replay_counter_base.py
- pcie_replay_counter_max.py
- pcie_replay_counter_min.py
- pcie_replay_counter_p1.py
- pcie_replay_counter_p10.py
- pcie_replay_counter_p25.py
- pcie_replay_counter_p5.py
- pcie_replay_counter_p50.py
- pcie_replay_counter_p75.py
- pcie_replay_counter_p90.py
- pcie_replay_counter_p95.py
- pcie_replay_counter_p99.py
- pcie_replay_counter_std.py
- pcie_transmit_throughput_avg.py
- pcie_transmit_throughput_base.py
- pcie_transmit_throughput_max.py
- pcie_transmit_throughput_min.py
- pcie_transmit_throughput_p1.py
- pcie_transmit_throughput_p10.py
- pcie_transmit_throughput_p25.py
- pcie_transmit_throughput_p5.py
- pcie_transmit_throughput_p50.py
- pcie_transmit_throughput_p75.py
- pcie_transmit_throughput_p90.py
- pcie_transmit_throughput_p95.py
- pcie_transmit_throughput_p99.py
- pcie_transmit_throughput_std.py
- power_throttle_duration_avg.py
- power_throttle_duration_base.py
- power_throttle_duration_max.py
- power_throttle_duration_min.py
- power_throttle_duration_p1.py
- power_throttle_duration_p10.py
- power_throttle_duration_p25.py
- power_throttle_duration_p5.py
- power_throttle_duration_p50.py
- power_throttle_duration_p75.py
- power_throttle_duration_p90.py
- power_throttle_duration_p95.py
- power_throttle_duration_p99.py
- power_throttle_duration_std.py
- request_count_avg.py
- request_goodput_avg.py
- request_latency_avg.py
- request_latency_base.py
- request_latency_max.py
- request_latency_min.py
- request_latency_p1.py
- request_latency_p10.py
- request_latency_p25.py
- request_latency_p5.py
- request_latency_p50.py
- request_latency_p75.py
- request_latency_p90.py
- request_latency_p95.py
- request_latency_p99.py
- request_latency_std.py
- request_throughput_avg.py
- retired_pages_dbe_avg.py
- retired_pages_dbe_base.py
- retired_pages_dbe_max.py
- retired_pages_dbe_min.py
- retired_pages_dbe_p1.py
- retired_pages_dbe_p10.py
- retired_pages_dbe_p25.py
- retired_pages_dbe_p5.py
- retired_pages_dbe_p50.py
- retired_pages_dbe_p75.py
- retired_pages_dbe_p90.py
- retired_pages_dbe_p95.py
- retired_pages_dbe_p99.py
- retired_pages_dbe_std.py
- retired_pages_sbe_avg.py
- retired_pages_sbe_base.py
- retired_pages_sbe_max.py
- retired_pages_sbe_min.py
- retired_pages_sbe_p1.py
- retired_pages_sbe_p10.py
- retired_pages_sbe_p25.py
- retired_pages_sbe_p5.py
- retired_pages_sbe_p50.py
- retired_pages_sbe_p75.py
- retired_pages_sbe_p90.py
- retired_pages_sbe_p95.py
- retired_pages_sbe_p99.py
- retired_pages_sbe_std.py
- sm_utilization_avg.py
- sm_utilization_base.py
- sm_utilization_max.py
- sm_utilization_min.py
- sm_utilization_p1.py
- sm_utilization_p10.py
- sm_utilization_p25.py
- sm_utilization_p5.py
- sm_utilization_p50.py
- sm_utilization_p75.py
- sm_utilization_p90.py
- sm_utilization_p95.py
- sm_utilization_p99.py
- sm_utilization_std.py
- thermal_throttle_duration_avg.py
- thermal_throttle_duration_base.py
- thermal_throttle_duration_max.py
- thermal_throttle_duration_min.py
- thermal_throttle_duration_p1.py
- thermal_throttle_duration_p10.py
- thermal_throttle_duration_p25.py
- thermal_throttle_duration_p5.py
- thermal_throttle_duration_p50.py
- thermal_throttle_duration_p75.py
- thermal_throttle_duration_p90.py
- thermal_throttle_duration_p95.py
- thermal_throttle_duration_p99.py
- thermal_throttle_duration_std.py
- time_to_first_token_avg.py
- time_to_first_token_base.py
- time_to_first_token_max.py
- time_to_first_token_min.py
- time_to_first_token_p1.py
- time_to_first_token_p10.py
- time_to_first_token_p25.py
- time_to_first_token_p5.py
- time_to_first_token_p50.py
- time_to_first_token_p75.py
- time_to_first_token_p90.py
- time_to_first_token_p95.py
- time_to_first_token_p99.py
- time_to_first_token_std.py
- time_to_second_token_avg.py
- time_to_second_token_base.py
- time_to_second_token_max.py
- time_to_second_token_min.py
- time_to_second_token_p1.py
- time_to_second_token_p10.py
- time_to_second_token_p25.py
- time_to_second_token_p5.py
- time_to_second_token_p50.py
- time_to_second_token_p75.py
- time_to_second_token_p90.py
- time_to_second_token_p95.py
- time_to_second_token_p99.py
- time_to_second_token_std.py
- total_ecc_dbe_aggregate_avg.py
- total_ecc_dbe_aggregate_base.py
- total_ecc_dbe_aggregate_max.py
- total_ecc_dbe_aggregate_min.py
- total_ecc_dbe_aggregate_p1.py
- total_ecc_dbe_aggregate_p10.py
- total_ecc_dbe_aggregate_p25.py
- total_ecc_dbe_aggregate_p5.py
- total_ecc_dbe_aggregate_p50.py
- total_ecc_dbe_aggregate_p75.py
- total_ecc_dbe_aggregate_p90.py
- total_ecc_dbe_aggregate_p95.py
- total_ecc_dbe_aggregate_p99.py
- total_ecc_dbe_aggregate_std.py
- total_ecc_dbe_volatile_avg.py
- total_ecc_dbe_volatile_base.py
- total_ecc_dbe_volatile_max.py
- total_ecc_dbe_volatile_min.py
- total_ecc_dbe_volatile_p1.py
- total_ecc_dbe_volatile_p10.py
- total_ecc_dbe_volatile_p25.py
- total_ecc_dbe_volatile_p5.py
- total_ecc_dbe_volatile_p50.py
- total_ecc_dbe_volatile_p75.py
- total_ecc_dbe_volatile_p90.py
- total_ecc_dbe_volatile_p95.py
- total_ecc_dbe_volatile_p99.py
- total_ecc_dbe_volatile_std.py
- total_ecc_sbe_aggregate_avg.py
- total_ecc_sbe_aggregate_base.py
- total_ecc_sbe_aggregate_max.py
- total_ecc_sbe_aggregate_min.py
- total_ecc_sbe_aggregate_p1.py
- total_ecc_sbe_aggregate_p10.py
- total_ecc_sbe_aggregate_p25.py
- total_ecc_sbe_aggregate_p5.py
- total_ecc_sbe_aggregate_p50.py
- total_ecc_sbe_aggregate_p75.py
- total_ecc_sbe_aggregate_p90.py
- total_ecc_sbe_aggregate_p95.py
- total_ecc_sbe_aggregate_p99.py
- total_ecc_sbe_aggregate_std.py
- total_ecc_sbe_volatile_avg.py
- total_ecc_sbe_volatile_base.py
- total_ecc_sbe_volatile_max.py
- total_ecc_sbe_volatile_min.py
- total_ecc_sbe_volatile_p1.py
- total_ecc_sbe_volatile_p10.py
- total_ecc_sbe_volatile_p25.py
- total_ecc_sbe_volatile_p5.py
- total_ecc_sbe_volatile_p50.py
- total_ecc_sbe_volatile_p75.py
- total_ecc_sbe_volatile_p90.py
- total_ecc_sbe_volatile_p95.py
- total_ecc_sbe_volatile_p99.py
- total_ecc_sbe_volatile_std.py
- total_gpu_memory_avg.py
- total_gpu_memory_base.py
- total_nvlink_crc_data_errors_avg.py
- total_nvlink_crc_data_errors_base.py
- total_nvlink_crc_data_errors_max.py
- total_nvlink_crc_data_errors_min.py
- total_nvlink_crc_data_errors_p1.py
- total_nvlink_crc_data_errors_p10.py
- total_nvlink_crc_data_errors_p25.py
- total_nvlink_crc_data_errors_p5.py
- total_nvlink_crc_data_errors_p50.py
- total_nvlink_crc_data_errors_p75.py
- total_nvlink_crc_data_errors_p90.py
- total_nvlink_crc_data_errors_p95.py
- total_nvlink_crc_data_errors_p99.py
- total_nvlink_crc_data_errors_std.py
- total_nvlink_crc_flit_errors_avg.py
- total_nvlink_crc_flit_errors_base.py
- total_nvlink_crc_flit_errors_max.py
- total_nvlink_crc_flit_errors_min.py
- total_nvlink_crc_flit_errors_p1.py
- total_nvlink_crc_flit_errors_p10.py
- total_nvlink_crc_flit_errors_p25.py
- total_nvlink_crc_flit_errors_p5.py
- total_nvlink_crc_flit_errors_p50.py
- total_nvlink_crc_flit_errors_p75.py
- total_nvlink_crc_flit_errors_p90.py
- total_nvlink_crc_flit_errors_p95.py
- total_nvlink_crc_flit_errors_p99.py
- total_nvlink_crc_flit_errors_std.py
- video_decoder_utilization_avg.py
- video_decoder_utilization_base.py
- video_decoder_utilization_max.py
- video_decoder_utilization_min.py
- video_decoder_utilization_p1.py
- video_decoder_utilization_p10.py
- video_decoder_utilization_p25.py
- video_decoder_utilization_p5.py
- video_decoder_utilization_p50.py
- video_decoder_utilization_p75.py
- video_decoder_utilization_p90.py
- video_decoder_utilization_p95.py
- video_decoder_utilization_p99.py
- video_decoder_utilization_std.py
- video_encoder_utilization_avg.py
- video_encoder_utilization_base.py
- video_encoder_utilization_max.py
- video_encoder_utilization_min.py
- video_encoder_utilization_p1.py
- video_encoder_utilization_p10.py
- video_encoder_utilization_p25.py
- video_encoder_utilization_p5.py
- video_encoder_utilization_p50.py
- video_encoder_utilization_p75.py
- video_encoder_utilization_p90.py
- video_encoder_utilization_p95.py
- video_encoder_utilization_p99.py
- video_encoder_utilization_std.py
- xid_last_error_avg.py
- xid_last_error_base.py
- xid_last_error_max.py
- xid_last_error_min.py
- xid_last_error_p1.py
- xid_last_error_p10.py
- xid_last_error_p25.py
- xid_last_error_p5.py
- xid_last_error_p50.py
- xid_last_error_p75.py
- xid_last_error_p90.py
- xid_last_error_p95.py
- xid_last_error_p99.py
- xid_last_error_std.py
- gpu_record.py
- record.py
- analyze.py
- common.py
- config.py
- process_export_files.py
- profile.py
- subcommand.py
- template.py
- __init__.py
- dcgm_telemetry_data_collector.py
- telemetry_data_collector.py
- triton_telemetry_data_collector.py
- .gitignore
- __init__.py
- constants.py
- exceptions.py
- logging.py
- main.py
- parser.py
- tokenizer.py
- types.py
- utils.py
- test_multiturn.py
- test_telemetry.py
- test_create_telemetry_data_collectors.py
- test_dcgm_telemetry_data_collector.py
- test_telemetry_data_collector.py
- test_triton_telemetry_data_collector.py
- test_dynamic_grpc_converter.py
- test_embeddings_converter.py
- test_huggingface_generate_converter.py
- test_image_retrieval_converter.py
- test_nvclip_converter.py
- test_openai_chat_converter.py
- test_openai_completions_converter.py
- test_output_format_converter.py
- test_rankings_converter.py
- test_template_converter.py
- test_tensorrtllm_engine_converter.py
- test_triton_generate_converter.py
- test_triton_tensorrtllm_converter.py
- test_triton_vllm_converter.py
- test_image_retrieval_profile_data_parser.py
- test_llm_profile_data_parser.py
- test_merged_profile_data_parser.py
- test_profile_data_parser.py
- test_console_exporter.py
- test_csv_exporter.py
- test_data_exporter_factory.py
- test_json_exporter.py
- test_telemetry_data_exporter_util.py
- test_llm_metrics.py
- test_metrics.py
- test_telemetry_metrics.py
- test_telemetry_stats_aggregator.py
- test_file_input_retriever.py
- test_input_retriever_factory.py
- test_payload_input_retriever.py
- test_synthetic_audio_generator.py
- test_synthetic_data_retriever.py
- test_synthetic_image_generator.py
- test_synthetic_prompt_generator.py
- test_statistics.py
- test_telemetry_statistics.py
- __init__.py
- test_artifacts.py
- test_base_config.py
- test_checkpoint.py
- test_cli.py
- test_config_command.py
- test_config_field.py
- test_genai_perf_config.py
- test_inputs.py
- test_library.py
- test_llm_goodput_calculator.py
- test_model_config_measurement.py
- test_perf_analyzer_config.py
- test_plot_configs.py
- test_process_export_files.py
- test_record.py
- test_results.py
- test_run_config.py
- test_run_config_measurement.py
- test_search_parameters.py
- test_subcommand.py
- test_sweep_objective_generator.py
- test_tokenizer.py
- test_utils.py
- .gitignore
- LICENSE
- pyproject.toml
- pytest.ini
- README.md
- CMakeLists.txt
- dynamic_grpc_client.cc
- dynamic_grpc_client.h
- dynamic_grpc_client_backend.cc
- dynamic_grpc_client_backend.h
- dynamic_grpc_infer_input.cc
- dynamic_grpc_infer_input.h
- CMakeLists.txt
- http_client.cc
- http_client.h
- openai_client.cc
- openai_client.h
- openai_client_backend.cc
- openai_client_backend.h
- openai_infer_input.cc
- openai_infer_input.h
- CMakeLists.txt
- CompileProto.cmake
- tfserve_client_backend.cc
- tfserve_client_backend.h
- tfserve_grpc_client.cc
- tfserve_grpc_client.h
- tfserve_infer_input.cc
- tfserve_infer_input.h
- CMakeLists.txt
- torchserve_client_backend.cc
- torchserve_client_backend.h
- torchserve_http_client.cc
- torchserve_http_client.h
- torchserve_infer_input.cc
- torchserve_infer_input.h
- CMakeLists.txt
- test_triton_client_backend.cc
- triton_client_backend.cc
- triton_client_backend.h
- alloc_payload.h
- c_api_infer_results.h
- CMakeLists.txt
- response_output.h
- scoped_defer.cc
- scoped_defer.h
- shared_library.cc
- shared_library.h
- shared_memory_manager.cc
- shared_memory_manager.h
- triton_c_api_backend.cc
- triton_c_api_backend.h
- triton_loader.cc
- triton_loader.h
- client_backend.cc
- client_backend.h
- CMakeLists.txt
- mock_client_backend.h
- __init__.py
- cli.py
- payload_dataset_manager.cc
- payload_dataset_manager.h
- payload_json_utils.cc
- payload_json_utils.h
- request_handler.cc
- request_handler.h
- response_json_utils.cc
- response_json_utils.h
- session_concurrency_manager.cc
- session_concurrency_manager.h
- base_queue_ctx_id_tracker.h
- CMakeLists.txt
- command_line_parser.cc
- command_line_parser.h
- concurrency_ctx_id_tracker.h
- concurrency_manager.cc
- concurrency_manager.h
- concurrency_worker.cc
- concurrency_worker.h
- constants.h
- ctx_id_tracker_factory.h
- cuda_runtime_library_manager.cc
- cuda_runtime_library_manager.h
- custom_load_manager.cc
- custom_load_manager.h
- custom_request_schedule_manager.cc
- custom_request_schedule_manager.h
- data_loader.cc
- data_loader.h
- doctest.h
- fifo_ctx_id_tracker.h
- ictx_id_tracker.h
- idle_timer.h
- iinfer_data_manager.h
- infer_context.cc
- infer_context.h
- infer_data.h
- infer_data_manager.cc
- infer_data_manager.h
- infer_data_manager_base.cc
- infer_data_manager_base.h
- infer_data_manager_factory.h
- infer_data_manager_shm.cc
- infer_data_manager_shm.h
- inference_load_mode.h
- inference_profiler.cc
- inference_profiler.h
- ischeduler.h
- iworker.h
- load_manager.cc
- load_manager.h
- load_worker.cc
- load_worker.h
- main.cc
- metrics.h
- metrics_manager.cc
- metrics_manager.h
- mock_concurrency_worker.h
- mock_data_loader.h
- mock_infer_context.h
- mock_infer_data_manager.h
- mock_inference_profiler.h
- mock_load_manager.h
- mock_model_parser.h
- mock_profile_data_collector.h
- mock_profile_data_exporter.h
- mock_request_rate_worker.h
- mock_sequence_manager.h
- model_parser.cc
- model_parser.h
- mpi_utils.cc
- mpi_utils.h
- perf_analyzer.cc
- perf_analyzer.h
- perf_analyzer_exception.h
- perf_analyzer_unit_tests.cc
- perf_utils.cc
- perf_utils.h
- periodic_concurrency_manager.cc
- periodic_concurrency_manager.h
- periodic_concurrency_worker.cc
- periodic_concurrency_worker.h
- profile_data_collector.cc
- profile_data_collector.h
- profile_data_exporter.cc
- profile_data_exporter.h
- rand_ctx_id_tracker.h
- rapidjson_utils.h
- rate_schedule.h
- report_writer.cc
- report_writer.h
- request_rate_manager.cc
- request_rate_manager.h
- request_rate_worker.cc
- request_rate_worker.h
- request_record.h
- sequence_manager.cc
- sequence_manager.h
- sequence_status.h
- tensor_data.h
- test_command_line_parser.cc
- test_concurrency_manager.cc
- test_ctx_id_tracker.cc
- test_custom_load_manager.cc
- test_custom_request_schedule_manager.cc
- test_dataloader.cc
- test_http_client.cc
- test_idle_timer.cc
- test_infer_context.cc
- test_inference_profiler.cc
- test_load_manager.cc
- test_load_manager_base.h
- test_metrics_manager.cc
- test_model_parser.cc
- test_payload_dataset_manager.cc
- test_payload_json_utils.cc
- test_perf_utils.cc
- test_profile_data_collector.cc
- test_profile_data_exporter.cc
- test_report_writer.cc
- test_request_rate_manager.cc
- test_response_json_utils.cc
- test_sequence_manager.cc
- test_utils.h
- thread_config.h
- thread_stat.h
- customizable_frontends_template
- embeddings_template
- example_plots_template
- files_template
- lora_template
- multi_modal_template
- rankings_template
- README_template
- tutorial_template
- version_template
- generate_docs.py
- template_vars.yaml
- add_copyright.py
- sharegpt_dataset_converter.py
- .clang-format
- .gitignore
- .pre-commit-config.yaml
- CMakeLists.txt
- LICENSE
- pyproject.toml
- README.md
- SECURITY.md
# Installation Guide
1. Get the code
git clone https://github.com/triton-inference-server/perf_analyzer
Downloads the entire project code from GitHub to your computer.
cd perf_analyzer
Moves into the project folder you just downloaded.
2. CMake
Medium RecommendedPrerequisites
mkdir build && cd build
Creates a folder to hold the build output and moves into it.
cmake ..
Analyzes the source code and generates build configuration files (must be run inside the build folder).
make
Compiles the code based on the generated build configuration to produce an executable.
Check that an executable was created inside the build folder, then run it directly (e.g. ./build/app_name).
3. Python
EasyPrerequisites
pip install .
Installs the package published on PyPI directly β no need to clone the source.
python <μ€νν νμΌλͺ
>.py # READMEμμ μ νν μ€ν νμΌλͺ
μ νμΈνμΈμ
Runs the Python script (or module).
If it runs without errors and prints output in the terminal, it worked.
// repository documentation
Was this content helpful?
(0 ratings)