{"id":"https://openalex.org/W7225627339","doi":"https://doi.org/10.5281/zenodo.24071325","title":"HyperMesh-512: Hyper-Parallel Inference Acceleration Architecture for Massive Language Models Based on a 2D Mesh NoC and a Decentralized 32-Channel LPDDR4X/DDR4 Bus","display_name":"HyperMesh-512: Hyper-Parallel Inference Acceleration Architecture for Massive Language Models Based on a 2D Mesh NoC and a Decentralized 32-Channel LPDDR4X/DDR4 Bus","publication_year":2026,"publication_date":"2026-11-09","ids":{"openalex":"https://openalex.org/W7225627339","doi":"https://doi.org/10.5281/zenodo.24071325"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.24071325","doi":"https://doi.org/10.5281/zenodo.24071325","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24071325","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.24071325","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137889222","display_name":"Tiago Barbosa Dias Maciel","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Tiago Barbosa Dias Maciel","raw_affiliation_strings":["Antigravity Advanced Materials Laboratory, tiago@brmg.com.br"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Antigravity Advanced Materials Laboratory, tiago@brmg.com.br","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5137889222"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10829","display_name":"Interconnection Networks and Systems","score":0.6011000275611877,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10829","display_name":"Interconnection Networks and Systems","score":0.6011000275611877,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.34860000014305115,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11522","display_name":"VLSI and FPGA Design Techniques","score":0.006099999882280827,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/autoregressive-decoding","display_name":"autoregressive decoding","score":0.8180000185966492},{"id":"https://openalex.org/keywords/memory-wall","display_name":"memory wall","score":0.8090000152587891},{"id":"https://openalex.org/keywords/chiplets","display_name":"chiplets","score":0.6179999709129333},{"id":"https://openalex.org/keywords/pcie","display_name":"PCI Express","score":0.5619999766349792},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"memory bandwidth","score":0.6589999794960022},{"id":"https://openalex.org/keywords/cost-of-capital","display_name":"cost of capital","score":0.3659999966621399}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7002000212669373},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5404999852180481},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.524399995803833},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3871999979019165},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.3711000084877014},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.320499986410141},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3199999928474426},{"id":"https://openalex.org/C87450203","wikidata":"https://www.wikidata.org/wiki/Q831940","display_name":"Operating expense","level":2,"score":0.31929999589920044},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.31610000133514404},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.31369999051094055},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.31220000982284546},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.31150001287460327},{"id":"https://openalex.org/C4216890","wikidata":"https://www.wikidata.org/wiki/Q815823","display_name":"Business model","level":2,"score":0.31130000948905945},{"id":"https://openalex.org/C188087704","wikidata":"https://www.wikidata.org/wiki/Q369577","display_name":"Standardization","level":2,"score":0.3093000054359436},{"id":"https://openalex.org/C154690210","wikidata":"https://www.wikidata.org/wiki/Q1668499","display_name":"Rewriting","level":2,"score":0.3086000084877014},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.3012999892234802},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.29269999265670776},{"id":"https://openalex.org/C144559511","wikidata":"https://www.wikidata.org/wiki/Q2986279","display_name":"Principal (computer security)","level":2,"score":0.29179999232292175},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.29010000824928284},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.28859999775886536},{"id":"https://openalex.org/C77390884","wikidata":"https://www.wikidata.org/wiki/Q217302","display_name":"Application-specific integrated circuit","level":2,"score":0.288100004196167},{"id":"https://openalex.org/C53833338","wikidata":"https://www.wikidata.org/wiki/Q1061424","display_name":"Context switch","level":2,"score":0.28600001335144043},{"id":"https://openalex.org/C2779602883","wikidata":"https://www.wikidata.org/wiki/Q15544750","display_name":"Memory architecture","level":2,"score":0.2736000120639801},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.26350000500679016},{"id":"https://openalex.org/C2779439359","wikidata":"https://www.wikidata.org/wiki/Q317088","display_name":"Commodity","level":2,"score":0.26330000162124634},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.2606000006198883},{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.25999999046325684},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.25450000166893005},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.25290000438690186}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.24071325","doi":"https://doi.org/10.5281/zenodo.24071325","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24071325","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},{"id":"doi:10.5281/zenodo.24071326","doi":"https://doi.org/10.5281/zenodo.24071326","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24071326","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.5281/zenodo.24071325","doi":"https://doi.org/10.5281/zenodo.24071325","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24071325","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7169211590","https://openalex.org/W4407693361","https://openalex.org/W7164931892","https://openalex.org/W4415317322","https://openalex.org/W7225229284","https://openalex.org/W4407308982","https://openalex.org/W4417114546","https://openalex.org/W4412964021","https://openalex.org/W7134111884","https://openalex.org/W7133515264"],"abstract_inverted_index":{"HyperMesh-512":[0],"is":[1],"a":[2,52,82,134,185],"hyper-parallel":[3],"PCIe":[4],"inference":[5],"accelerator":[6],"architecture":[7],"engineered":[8],"to":[9,47,64,87,99,108],"overcome":[10],"the":[11,59],"Memory":[12],"Wall":[13],"bottleneck":[14],"during":[15],"autoregressive":[16],"decoding":[17],"of":[18,34,67],"massive":[19],"language":[20],"models":[21],"(such":[22],"as":[23],"Llama":[24],"3.1":[25],"70B/405B":[26],"and":[27,102,138],"DeepSeek-R1/V3":[28],"671B":[29],"MoE).":[30],"Featuring":[31],"512":[32],"GB":[33,97,101],"commodity":[35],"LPDDR4X/DDR4":[36],"memory":[37,69],"distributed":[38],"across":[39],"32":[40],"independent":[41],"physical":[42],"64-bit":[43],"point-to-point":[44],"channels":[45],"coupled":[46],"16":[48],"ASIC":[49],"micro-chiplets":[50],"in":[51],"2D":[53],"Mesh":[54],"NoC":[55],"(X-Y":[56],"DOR":[57],"routing),":[58],"board":[60],"delivers":[61],"819.2":[62],"GB/s":[63,66,110],"1,092.1":[65],"sustained":[68],"bandwidth.":[70],"To":[71],"address":[72],"ultra-long":[73],"context":[74,105],"windows":[75],"(\u2265":[76],"128k":[77,92],"tokens),":[78],"each":[79],"chiplet":[80,198],"integrates":[81],"dedicated":[83],"single-cycle":[84],"affine":[85],"INT4":[86],"INT16":[88],"hardware":[89],"decompressor,":[90],"reducing":[91],"KV-cache":[93],"footprint":[94],"from":[95],"41.94":[96],"down":[98],"10.49":[100],"raising":[103],"effective":[104],"read":[106],"bandwidth":[107],"3,276.8":[109],"equivalent.":[111],"Audited":[112],"Industrial":[113],"BOM:":[114],"US$":[115,119,190,195],"1,680.00.":[116],"Target":[117],"MSRP:":[118],"4,999.00":[120],"delivering":[121],"102.42":[122],"GB/k$":[123],"(a":[124],"48x":[125],"capital-efficiency":[126],"advantage":[127],"over":[128],"corporate":[129],"HGX":[130],"H100":[131],"clusters),":[132],"with":[133],"98.33%":[135],"CapEx":[136],"reduction":[137,142],"96.57%":[139],"electrical":[140],"OpEx":[141],"for":[143],"on-premise,":[144],"air-gapped":[145],"private":[146],"generative":[147],"AI":[148],"workloads.":[149],"Declara\u00e7\u00e3o":[150],"de":[151,173,189],"Natureza":[152],"Te\u00f3rica:":[153],"Artigo":[154],"te\u00f3rico":[155],"/":[156],"preprint":[157],"t\u00e9cnico":[158],"contendo":[159],"modelagem":[160],"arquitetural,":[161],"RTL":[162],"sintetiz\u00e1vel":[163],"e":[164,169,187],"proje\u00e7\u00f5es":[165],"tecno-econ\u00f4micas.":[166],"Propriedade":[167],"Intelectual":[168],"Licenciamento":[170],"(Royalties):":[171],"Titularidade":[172],"Tiago":[174],"Barbosa":[175],"Dias":[176],"Maciel":[177],"(CPF":[178],"060.184.906-09,":[179],"tiago@brmg.com.br).":[180],"Qualquer":[181],"explora\u00e7\u00e3o":[182],"comercial":[183],"sujeita":[184],"licenciamento":[186],"royalty":[188],"50,00":[191],"por":[192,197],"placa":[193],"ou":[194],"3,00":[196],"integrado.":[199]},"counts_by_year":[],"updated_date":"2026-11-11T12:16:26.678958","created_date":"2026-11-10T00:00:00"}
