[{"data":1,"prerenderedAt":37},["ShallowReactive",2],{"post-document:\u002Fdeep_learning\u002Ftechnics\u002F2026\u002F08\u002F06\u002Fdistributed-training-mlops\u002F":3},{"id":4,"title":5,"body":6,"categories":17,"date":20,"description":21,"extension":22,"image":23,"key_concepts":23,"last_modified_at":23,"legacyPath":24,"meta":25,"navigation":27,"part":23,"path":28,"published":27,"robots":23,"seo":29,"series":23,"stem":30,"strengths":23,"summary":31,"tags":32,"tradeoffs":23,"__hash__":36},"posts\u002Fposts\u002FDeep_Learning\u002FTechnics\u002F2026-08-06-distributed-training-mlops.md","분산 학습과 MLOps",{"type":7,"value":8,"toc":13},"minimark",[9],[10,11,12],"p",{},"Single GPU와 Multi-GPU\nData Parallelism\nDistributed Data Parallel\nModel Parallelism\nTensor Parallelism\nPipeline Parallelism\nFully Sharded Data Parallel\nZeRO Optimization\nGradient Checkpointing\nDistributed Checkpoint\nDeepSpeed\n실험 관리\n모델 및 데이터 버전 관리\n학습 로그 관리\n배포 후 성능 모니터링\nData Drift와 Model Drift\n재학습 파이프라인\nA\u002FB Test\nCanary Deployment\n모델 롤백",{"title":14,"searchDepth":15,"depth":15,"links":16},"",2,[],[18,19],"Deep_Learning","Technics","2026-08-06 00:00:00 +0900","멀티 GPU 분산 학습 방식과 실험 관리, 모델 배포, 모니터링, 재학습 파이프라인을 정리한다.","md",null,"\u002Fdeep_learning\u002Ftechnics\u002F2026\u002F08\u002F06\u002Fdistributed-training-mlops\u002F",{"layout":26},"post",true,"\u002Fposts\u002Fdeep_learning\u002Ftechnics\u002F2026-08-06-distributed-training-mlops",{"title":5,"description":21},"posts\u002FDeep_Learning\u002FTechnics\u002F2026-08-06-distributed-training-mlops","딥러닝 분산 학습 기술과 운영 단계의 MLOps 핵심 항목을 정리한다.",[33,34,35],"Deep Learning","Distributed Training","MLOps","iZ-3UsAFM71KxtpWA2V-CY98YMyOrwFZgS0AE47U1fg",1788744789086]