{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/defines-enabling-fast-exploration-of-the","title":"DeFiNES: Enabling Fast Exploration of the Depth-first Scheduling Space for DNN Accelerators through Analytical Modeling","arxiv_id":"2212.05344","date":"2022-12-10","proceeding":null,"authors":["Linyan Mei","Koen Goetschalckx","Arne Symons","Marian Verhelst"],"abstract":"DNN workloads can be scheduled onto DNN accelerators in many different ways: from layer-by-layer scheduling to cross-layer depth-first scheduling (a.k.a. layer fusion, or cascaded execution). This results in a very broad scheduling space, with each schedule leading to varying hardware (HW) costs in terms of energy and latency. To rapidly explore this vast space for a wide variety of hardware architectures, analytical cost models are crucial to estimate scheduling effects on the HW level. However, state-of-the-art cost models are lacking support for exploring the complete depth-first scheduling space, for instance focusing only on activations while ignoring weights, or modeling only DRAM accesses while overlooking on-chip data movements. These limitations prevent researchers from systematically and accurately understanding the depth-first scheduling space. After formalizing this design space, this work proposes a unified modeling framework, DeFiNES, for layer-by-layer and depth-first scheduling to fill in the gaps. DeFiNES enables analytically estimating the hardware cost for possible schedules in terms of both energy and latency, while considering data access at every memory level. This is done for each schedule and HW architecture under study by optimally choosing the active part of the memory hierarchy per unique combination of operand, layer, and feature map tile. The hardware costs are estimated, taking into account both data computation and data copy phases. The analytical cost model is validated against measured data from a taped-out depth-first DNN accelerator, DepFiN, showing good modeling accuracy at the end-to-end neural network level. A comparison with generalized state-of-the-art demonstrates up to 10X better solutions found with DeFiNES.","url_abs":"https://arxiv.org/abs/2212.05344v4","url_pdf":"https://arxiv.org/pdf/2212.05344v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"links_only","authors_date_abstract":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license), from the Kaggle arXiv metadata snapshot of 2026-09-12"},"code_links":[{"paper_slug":"defines-enabling-fast-exploration-of-the","repo_url":"https://github.com/zigzag-project/defines","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"defines-enabling-fast-exploration-of-the","repo_url":"https://github.com/ZigZag-Project/zigzag","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"defines-enabling-fast-exploration-of-the","repo_url":"https://github.com/kuleuven-micas/acco","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2212.05344","atlas_url":"https://app.syntology.ai/?focus=2212.05344","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2212.05344"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/zigzag-project/defines","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ZigZag-Project/zigzag","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kuleuven-micas/acco","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"summary":{"ran":6,"unverified":3},"by_repo_kind":{"official":{"samples":2,"ran":2,"repositories":1},"listed":{"samples":7,"ran":4,"repositories":2}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"270a7a067a1c0319","entry":"annotate_heatmap","repo":"zigzag-project/defines","repo_kind":"official","path":"plot_helper_funcs.py","file_url":"https://github.com/zigzag-project/defines/blob/HEAD/plot_helper_funcs.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"270a7a067a1c0319"}},{"code_sha256_prefix":"528110165dfbbbed","entry":"extract_port_latency","repo":"kuleuven-micas/acco","repo_kind":"listed","path":"classes/depthfirst/data_copy_layer.py","file_url":"https://github.com/kuleuven-micas/acco/blob/HEAD/classes/depthfirst/data_copy_layer.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"528110165dfbbbed"}},{"code_sha256_prefix":"fe45575f66dbe74b","entry":"hash_sha512","repo":"ZigZag-Project/zigzag","repo_kind":"listed","path":"zigzag/utils.py","file_url":"https://github.com/ZigZag-Project/zigzag/blob/HEAD/zigzag/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fe45575f66dbe74b"}},{"code_sha256_prefix":"cfe5673960d811ad","entry":"heatmap","repo":"zigzag-project/defines","repo_kind":"official","path":"plot_helper_funcs.py","file_url":"https://github.com/zigzag-project/defines/blob/HEAD/plot_helper_funcs.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"cfe5673960d811ad"}},{"code_sha256_prefix":"979d8736de4592ce","entry":"pickle_deepcopy","repo":"ZigZag-Project/zigzag","repo_kind":"listed","path":"zigzag/utils.py","file_url":"https://github.com/ZigZag-Project/zigzag/blob/HEAD/zigzag/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"979d8736de4592ce"}},{"code_sha256_prefix":"c54085ffaf8bf13f","entry":"pickle_deepcopy","repo":"kuleuven-micas/acco","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/kuleuven-micas/acco/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"c54085ffaf8bf13f"}},{"code_sha256_prefix":"9247aaa762351759","entry":"calc_MUW_union","repo":"kuleuven-micas/acco","repo_kind":"listed","path":"classes/cost_model/cost_model.py","file_url":"https://github.com/kuleuven-micas/acco/blob/HEAD/classes/cost_model/cost_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"9247aaa762351759"}},{"code_sha256_prefix":"f133edb257d3344c","entry":"get_shared_mem_list","repo":"kuleuven-micas/acco","repo_kind":"listed","path":"classes/cost_model/cost_model.py","file_url":"https://github.com/kuleuven-micas/acco/blob/HEAD/classes/cost_model/cost_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"f133edb257d3344c"}},{"code_sha256_prefix":"10a1b5d040c1c4cb","entry":"pickle_save","repo":"ZigZag-Project/zigzag","repo_kind":"listed","path":"zigzag/utils.py","file_url":"https://github.com/ZigZag-Project/zigzag/blob/HEAD/zigzag/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"10a1b5d040c1c4cb"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}