Update mcore parallelism initialization in nemo2 #5999

Run time

Learn about OS pricing on GitHub Actions

Job	Run time
pre-flight	1s
cicd-cluster-clean	3s
gpu-test	16s
cicd-test-container-setup	3m 5s
L0_Unit_Tests_GPU_ASR / main	13m 15s
L0_Unit_Tests_GPU_Audio / main	2m 45s
L0_Unit_Tests_GPU_Common / main	2m 49s
L0_Unit_Tests_GPU_Hydra / main	2m 20s
L0_Unit_Tests_GPU_NLP / main	4m 12s
L0_Unit_Tests_GPU_LLM / main	2m 45s
L0_Unit_Tests_GPU_Multimodal / main	1m 41s
L0_Unit_Tests_CPU_ASR / main	7m 30s
OPTIONAL_L0_Unit_Tests_GPU_Core / main	21m 15s
L0_Unit_Tests_GPU_TTS / main	1m 38s
L0_Unit_Tests_CPU_Audio / main	59s
L0_Unit_Tests_CPU_Common / main	1m 18s
L0_Unit_Tests_CPU_Core / main	2m 49s
L0_Unit_Tests_CPU_Hydra / main	1m 0s
L0_Unit_Tests_CPU_LLM / main	37s
L0_Unit_Tests_GPU_Lightning / main	2m 53s
L0_Unit_Tests_GPU_Others / main	1m 36s
L0_Unit_Tests_CPU_Lightning / main	48s
L0_Unit_Tests_CPU_Multimodal / main	40s
L0_Unit_Tests_CPU_NLP / main	2m 58s
L0_Setup_Test_Data_And_Models / main	1m 21s
L0_Unit_Tests_CPU_Others / main	27s
L0_Unit_Tests_CPU_TTS / main	1m 47s
L2_Community_LLM_Checkpoints_tests_Bert / main	1m 53s
L2_Community_LLM_Checkpoints_tests_Llama / main	1m 49s
L2_Community_LLM_Checkpoints_tests_Llama3 / main	1m 50s
L2_Community_LLM_Checkpoints_tests_Mamba2 / main	1m 56s
L2_Community_LLM_Checkpoints_tests_Falcon / main	1m 44s
L2_Community_LLM_Checkpoints_tests_StarCoder / main	1m 44s
L2_Community_vita_Checkpoints_tests_Llama3 / main	5m 3s
L2_Distill_Llama2 / main	2m 58s
L2_Prune_Width_Llama2 / main	2m 24s
L2_PTQ_Llama2_Export_Only / main	1m 39s
L2_PTQ_Llama2_FP8 / main	1m 52s
ASR_dev_run_Speech_Pre-training_-_CitriNet / main	53s
ASR_dev_run_Speech_to_Text / main	58s
ASR_dev_run_Speech_To_Text_Finetuning / main	55s
ASR_dev_run_Speech_to_Text_WPE_-_CitriNet / main	58s
ASR_dev_run_Speech_To_Text_HF_Finetuning / main	1m 35s
ASR_dev_run_Speech_to_Text_WPE_-_Conformer / main	43s
ASR_dev_run-part_two_Speech_to_Text_WPE_-_Squeezeformer / main	42s
L2_ASR_Multi-dataloader_dev_run_Speech_to_Label_multi-dataloader / main	41s
L2_ASR_Multi-dataloader_dev_run_Speech_to_Text_multi-dataloader / main	54s
L2_PTQ_Llama2_INT8_SQ / main	1m 43s
L2_Speaker_dev_run_Speaker_Diarization / main	43s
L2_Speaker_dev_run_Speaker_Recognition / main	39s
L2_Speech_to_Text_AED / main	50s
L2_Speech_to_Text_EMA / main	1m 56s
L2_Speaker_dev_run_Clustering_Diarizer_Inference / main	1m 44s
L2_Speaker_dev_run_Multispeaker_ASR_Data_Simulation / main	1m 33s
L2_Speaker_dev_run_Neural_Diarizer_Inference / main	1m 46s
L2_Speaker_dev_run_Speaker_Diarization_with_ASR_Inference / main	1m 34s
L2_Speaker_dev_run_Speech_to_Label / main	40s
L2_ASR_Adapters_Linear_Adapters / main	44s
L2_ASR_Adapters_RelPos_MHA_Adapters / main	48s
L2_Speech_Batch_Size_OOMptimizer / main	2m 41s
L2_Speech_Batch_Size_OOMptimizer_Canary / main	2m 7s
L2_Speech_Estimate_Duration_Bins / main	1m 52s
L2_G2P_Models_G2P_Conformer_training_evaluation_and_inference / main	2m 1s
L2_Speech_Transcription_Canary_Transcribe_Full_Manifest / main	3m 30s
L2_Speech_Transcription_Canary_Transcribe_With_Prompt / main	3m 32s
L2_Speech_Transcription_Speech_to_Text_Transcribe / main	1m 35s
L2_G2P_Models_HeteronymClassificationModel_training_evaluation_and_inference / main	2m 38s
L2_NMT_Attention_is_All_You_Need_Finetuning / main	1m 46s
L2_NMT_Attention_is_All_You_Need_Inference / main	1m 54s
L2_NMT_Attention_is_All_You_Need_Training_NMT_Multi-Validation / main	1m 8s
L2_NMT_Attention_is_All_You_Need_Training_NMT_Training_Post-LN / main	1m 15s
L2_NMT_Attention_is_All_You_Need_Training_NMT_Training_Pre-LN / main	59s
L2_Segmentation_Tool_Parallel_ctc_segmentation_test_L2_Eng_CitriNet_with_wav / main	3m 24s
L2_Segmentation_Tool_Parallel_ctc_segmentation_test_L2_Ru_QN_with_mp3 / main	2m 59s
L2_Speech_Transcription_Canary_Transcribe_Audio_Dir / main	3m 36s
L2_Megatron_Bert_Pretraining_and_Resume_Training_with_Pipeline_Parallelism / main	3m 43s
L2_Megatron_Core_Bert_Pretraining_and_Resume_Training / main	3m 44s
L2_Megatron_NMT_Training_TP2 / main	4m 2s
L2_NMT_Tarred_Dataset_Creation_Script_Tarred_Dataset_Creation / main	1m 46s
L2_NMT_Tarred_Dataset_Creation_Auto_Tarred_Dataset_Creation / main	53s
L2_Pretraining_BERT_from_Preprocessed / main	54s
L2_Pretraining_BERT_pretraining_from_Text / main	51s
L2_Megatron_GPT_Pretraining_and_Resume_Training_TP2 / main	3m 28s
L2_Megatron_GPT_Skip_Train / main	2m 49s
L2_Megatron_GPT_with_ALiBi_Pretraining_and_Resume_Training_TP2 / main	3m 1s
L2_RAG_Pipeline_Generating / main	2m 25s
L2_RAG_Pipeline_Indexing / main	1m 55s
L2_Megatron_GPT_Finetuning_PP2 / main	4m 37s
L2_Megatron_GPT_with_Drop_Optimizer_States_TP2 / main	3m 23s
L2_Megatron_GPT_with_ResetLR_Pretraining_and_Resume_Training_TP2 / main	4m 11s
L2_Megatron_GPT_with_Rope_Pretraining_and_Resume_Training_TP2 / main	3m 0s
L2_Megatron_GPT_Embedding / main	2m 19s
L2_Megatron_GPT_Eval / main	1m 58s
L2_Megatron_GPT_Finetuning_StarCoder_PP1 / main	1m 23s
L2_Megatron_GPT_with_KERPLE_Pretraining_and_Resume_Training_TP2 / main	3m 1s
OPTIONAL_L2_Megatron_GPT_Auto_Configurator_TP1_PP1_MBS124 / main	2m 34s
Optional_L2_Megatron_GPT_Pretraining_and_Resume_Training_PP2 / main	3m 52s
L2_Megatron_GPT_Eval_PP2 / main	2m 36s
L2_Megatron_GPT_PEFT_Lora_PP2_O2 / main	2m 45s
L2_Megatron_GPT_PEFT_Lora_TP2_O1 / main	2m 46s
L2_Megatron_Change_Partitions_Increase_TP_Num_Partitions_-2_to_4-_and_PP_Num_Partitions_-1_to_2 / main	2m 2s
L2_Megatron_Core_T5_Pretraining_and_Resume_Training_TP2 / main	3m 19s
L2_Megatron_GPT_PEFT_Lora_TP2SP1 / main	1m 9s
L2_Megatron_Change_Partitions_Reduce_TP_Num_Partitions_-2_to_1-_and_PP_Num_Partitions_-1_to_2 / main	2m 1s
L2_Megatron_GPT_Reranker / main	1m 47s
L2_Megatron_Core_T5_Eval / main	1m 55s
L2_Megatron_GPT_SFT_Eval_inference_seq_len_greaterThan_training_seq_len / main	1m 59s
L2_Megatron_T5_w_Mixture_of_Expert_Pretraining / main	2m 5s
L2_Megatron_T5_with_ALiBi_Pretraining_and_Resume_Training_TP2 / main	2m 57s
L2_Megatron_T5_with_KERPLE_Pretraining_and_Resume_Training_TP2 / main	2m 53s
L2_Megatron_Core_T5_PEFT_Lora_TP2 / main	3m 39s
L2_Megatron_Mock_Data_Generation_MockGPTDataset / main	3m 4s
L2_Megatron_Mock_Data_Generation_MockT5Dataset / main	1m 55s
L2_Megatron_T5_Pretraining_and_Resume_Training_PP2 / main	2m 49s
L2_Megatron_UL2_Pretraining_and_Resume_Training_TP2 / main	2m 58s
L2_Stable_Diffusion_Training / main	4m 15s
L2_NeMo_2_GPT_DDP_Param_Parity_check / main	4m 38s
L2_TTS_Fast_dev_runs_1_FastPitch / main	1m 59s
L2_TTS_Fast_dev_runs_1_Mixer-TTS / main	2m 1s
L2_TTS_Fast_dev_runs_1_Hifigan / main	1m 45s
L2_TTS_Fast_dev_runs_1_Tacotron_2 / main	1m 4s
L2_TTS_Fast_dev_runs_1_WaveGlow / main	1m 43s
L2_NeMo_2_GPT_Pretraining_no_transformer_engine / main	3m 40s
L2_NeMo_2_HF_MODEL_IMPORT / main	4m 54s
L2_NeMo_2_SSM_Finetuning / main	4m 28s
L2_NeMo_2_llama3_pretraining_recipe / main	4m 37s
L2_NeMo_2_SSM_Pretraining / main	3m 45s
L2_NeMo_2_T5_Finetuning / main	4m 7s
L2_NeMo_2_T5_LoRA / main	4m 8s
L2_NeMo_2_T5_Pretraining / main	5m 30s
L2_NeMo_2_GPT_LoRA_TP1PP1_MBS1 / main	3m 38s
L2_NeMo_2_GPT_LoRA_TP1PP1_MBS2 / main	3m 39s
Speech_Checkpoints_tests / main	3m 46s
L2_NeMo_2_GPT_LoRA_TP1PP2_MBS2 / main	3m 34s
L2_NeMo_2_GPT_SFT_TP1PP1_MBS1 / main	4m 14s
L2_NeMo_2_GPT_SFT_TP1PP1_MBS1_PACKED / main	4m 9s
L2_NeMo_2_GPT_SFT_TP1PP1_MBS2 / main	4m 6s
L2_NeMo_2_GPT_SFT_TP1PP2_MBS2 / main	4m 4s
L2_NeMo_2_GPT_SFT_TP2PP1_MBS2 / main	4m 8s
L2_NeMo_2_Mixtral_Pretraining / main	2m 1s
L2_NeMo_2_GPT_LoRA_TP1PP1_MBS1_PACKED / main	3m 36s
L2_NeMo_2_GPT_LoRA_TP2PP1_MBS2 / main	3m 40s
L2_NeMo_2_NeMo_Mcore_Mixtral_bitexact / main	2m 53s
L2_NeMo_2_PTQ_Llama2_FP8 / main	3m 1s
Nemo_CICD_Test	4s
	6h 15m 11s

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update mcore parallelism initialization in nemo2 #5999

Usage

Update mcore parallelism initialization in nemo2 #5999

Jobs

Run details

Run time