{"corpus_id":276118480,"paper_sha":"e2996bc521adbdc49dffcdc627395a27290966e6","doi":"10.1109/ICITR64794.2024.10857753","arxiv_id":null,"pmid":null,"pmcid":null,"mag_id":null,"dblp_id":null,"acl_id":null,"title":"Distributed Deep Neural Networks Training in a Multi-Worker Environment","year":2024,"publication_date":"2024-12-05","venue":"2024 9th International Conference on Information Technology Research (ICITR)","journal":{"name":"2024 9th International Conference on Information Technology Research (ICITR)","pages":"1-6","volume":null},"journal_issn":null,"journal_title":null,"publication_types":["Conference"],"pubmed_pub_types":null,"s2_fields_of_study":[],"reference_count":10,"citation_count":0,"influential_citation_count":0,"is_open_access":false,"arxiv_categories":null,"arxiv_license":null,"arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":null,"s2_open_access_landing_url":null,"s2_open_access_license":null,"s2_open_access_status":null,"pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"Deep learning has become promising across numerous fields in transforming conventional paradigms into smart eras in distributed applications. Large neural networks in recent years have been popular in solving massive real-world problems. However, the challenge behind the increasing complexity of deep neural networks impacts the training time. Appropriate resource provisioning and rightsizing is the requirement in all standard platforms like the cloud to handle this performance degradation. This research explores distributed CPU clusters as a scalable and cost-effective alternative for training large neural networks. The experiments on two different multi-processing machines with workers' distributions demonstrated the change in maximum accuracies is in a range of 92.96% to 96.74%. As our approach can be adopted and experiments can be extended to serverful and serverless computing training workloads, deep learning researchers and practitioners will benefit from our solution.","claims":[{"public_id":"cl_42303760797b9f1374a1eeb6293fd211","status":"active","text":"Distributed CPU clusters are presented as a scalable and cost-effective alternative for training large neural networks.","confidence":0.93,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_42303760797b9f1374a1eeb6293fd211"},{"public_id":"cl_a53c5bc7dcdb21413824273584bca71d","status":"active","text":"Experiments on two multi-processing machines with different worker distributions yielded maximum accuracies ranging from 92.96% to 96.74%.","confidence":0.97,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_a53c5bc7dcdb21413824273584bca71d"},{"public_id":"cl_95c0c68bb12a4668fc8c8193d5acfbfd","status":"active","text":"The approach is intended to be adaptable to serverful and serverless computing training workloads.","confidence":0.81,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_95c0c68bb12a4668fc8c8193d5acfbfd"}],"concepts":[{"public_id":"co_167554f8d0a9a93814d78c4161c21583","status":"active","name":"worker distributions","description":"Different allocations of workers across the machines used in the distributed training setup.","types":["configuration"],"aliases":["distribution of workers"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_167554f8d0a9a93814d78c4161c21583"},{"public_id":"co_46167eae67acb75b6adc711f1ec2e5f6","status":"active","name":"maximum accuracies","description":"The highest accuracy values achieved during the reported experiments.","types":["evaluation metric"],"aliases":["max accuracies"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_46167eae67acb75b6adc711f1ec2e5f6"},{"public_id":"co_686ba6a3a5725ce174701431b5e67522","status":"active","name":"serverless computing","description":"A computing model that executes workloads without direct server management by the user.","types":["computing model"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_686ba6a3a5725ce174701431b5e67522"},{"public_id":"co_6dad5e2574b54a1ee1fc0583429e154e","status":"active","name":"training workloads","description":"Computational workloads associated with model training.","types":["workload"],"aliases":["training job workloads"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_6dad5e2574b54a1ee1fc0583429e154e"},{"public_id":"co_94701b74a3af116b2e0e5d286111e5b4","status":"active","name":"distributed CPU clusters","description":"Groups of CPU-based machines used together to execute training workloads in a distributed manner.","types":["computing infrastructure","method"],"aliases":["CPU clusters"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_94701b74a3af116b2e0e5d286111e5b4"},{"public_id":"co_9576d53197d3d64723f212d8c44fb3cc","status":"active","name":"multi-processing machines","description":"Computing machines configured to run multiple processes for distributed experimentation.","types":["computing platform"],"aliases":["multiprocessing machines"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_9576d53197d3d64723f212d8c44fb3cc"},{"public_id":"co_bea3b3de08c8c29807b4772271473355","status":"active","name":"serverful computing","description":"A computing model that runs workloads on provisioned servers or virtual machines.","types":["computing model"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_bea3b3de08c8c29807b4772271473355"},{"public_id":"co_c6d658f3a43276c20daa152f1bf69f43","status":"active","name":"training large neural networks","description":"The process of optimizing large neural network models using training data.","types":["task"],"aliases":["large neural network training"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_c6d658f3a43276c20daa152f1bf69f43"}],"external_ids":{"DOI":"10.1109/ICITR64794.2024.10857753","ArXiv":null,"PubMed":null,"PubMedCentral":null,"MAG":null,"DBLP":null,"ACL":null},"open_access":{"is_open_access":false,"pdf_url":null,"landing_url":"https://sah.borca.ai/papers/276118480","source":null,"pdf_url_source":null,"license":null,"reason":"pdf_url_not_indexed"},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":false,"full_text_source":null,"count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":633436,"paper_uid":"2e2e214d-f2f3-4ced-a486-9512a889d578","canonical_identity":{"paper_id":633436,"paper_uid":"2e2e214d-f2f3-4ced-a486-9512a889d578","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/276118480"}