[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81655-en":3,"doc-seo-81655-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},81655,16904993612988,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Co-Design Optimization for Data Center Cooling System via Digital Twin","Liquid-cooled exascale supercomputers remove heat using cooling plants with multiple parallel subloops, yet scalable methods for allocating coolant distribution units across subloops and distributing flow among them remain insufficient. A three-layer optimization framework jointly decides integer CDU partitioning, continuous flow-fraction allocation, and per-timestep co-design of total flow rate and supply temperature under subloop thermal safety constraints. A Modelica-based surrogate study evaluates 611 feasible partitions using Frontier full-year data.","arXiv :2605 . 15516v2 [ ee ss . SY] 10 Jul 2026  \nCO-DESIGN OPTIMIZATION FOR DATA CENTER COOLING  \nSYSTEM VIA DIGITAL TWIN  \nA PREPRINT  \nShrenik Jadhav  \nDepartment of Computer and Information Science  \nUniversity of Michigan-Dearborn  \n4901 Evergreen Rd, Dearborn, MI, USA  \nZheng Liu*  \nDepartment of Industrial and Manufacturing Systems Engineering  \nUniversity of Michigan-Dearborn  \n4901 Evergreen Rd, Dearborn, MI, USA  \n[zhengtl@umich.edu](zhengtl@umich.edu)  \nJuly 13, 2026  \nABSTRACT  \nLiquid-cooled exascale supercomputers dissipate heat through cooling plants organized as multiple parallel subloops, but how to allocate coolant distribution units (CDUs) across subloops and how to distribute flow among them has not been systematically addressed for facilities at this scale. This paper presents a three-layer optimization framework that jointly determines the integer partition of CDUs across subloops, the continuous flow fraction allocation, and the per-timestep co-design optimization of total flow rate and supply temperature subject to per-subloop thermal safety constraints. The Modelica simulation model is built based on the data of Frontier exascale supercomputer at Oak Ridge National Laboratory. By developing a reduced-order surrogate model, all 611 feasible partitions of 25 CDUs are evaluated across the full year operational dataset of 49,353 timesteps. Three progressively richer operational strategies are compared, ranging from flow control optimization to full three-layer co-design optimization with dynamically adjusted flow fractions. The optimal design within the surrogate optimization problem is a two-subloop plant achieving 35.48% annual cooling energy savings, only 0.18% above the current three-subloop Frontier design at 35.30% . Most of the savings are delivered by supervisory co-optimization of total flow rate and supply temperature; the distinct role of flow fraction optimization is design robustness rather than additional raw savings. Flow fraction optimization compensates for any feasible CDU-to-subloop assignment, reducing the design sensitivity by 93% and providing a low-cost software-only pathway to near-optimal performance on the existing Frontier hardware. The framework is transferable to other liquid-cooled high-performance computing plants.  \nKeywords Data center · thermal management · co-design optimization · digital twin  \n1 Introduction  \nData center electricity consumption has emerged as one of the most rapidly growing loads on the global power system. Worldwide data center demand reached 415 TWh in 2024, approximately 1.5% of total electricity use, and is projected to exceed 945 TWh by 2030 [1] . In the United States, data centers consumed 176 TWh in 2023, about 4.4% of national electricity, with projections ranging from 325 to 580 TWh by 2028 [2] . Cooling infrastructure accounts for 30 to  \n∗ Corresponding author.  \n40% of total facility electricity consumption [3, 4], making it the single largest controllable load in most data centers. Despite a decade of advances in efficient server hardware and air-side economization, the global average Power Usage Effectiveness has remained between 1.55 and 1.59 since 2020 [5], indicating that incremental improvements to existing cooling systems are approaching diminishing returns and that more systematic optimization approaches are needed.  \nHigh-performance computing (HPC) facilities experience this challenge at an amplified scale. Exascale supercomputers such as Frontier at Oak Ridge National Laboratory operate at power levels of 8 to 30 MW, employ 100% direct liquid cooling with variable-speed pumps, and exhibit rapid thermal transients driven by workload dynamics [6, 7] . These systems differ structurally from enterprise data centers in three important ways. First, liquid cooling loops respond to load changes on time scales of minutes rather than the seconds typical of forced-air cooling, introducing tight coupling between thermal inertia and actuator ","cbCaidW2JPvO3IDN","https://ap.wps.com/l/cbCaidW2JPvO3IDN","pdf",1042068,6,1,20,"English","en",105,"# Abstract\n# Introduction\n## Data center energy impact and cooling as a dominant load\n## Exascale HPC cooling characteristics and design complexity\n## Limitations of existing Frontier-focused work\n## Physics-based digital twins as a direction","[{\"question\":\"What problem does the paper address for exascale data center cooling plants?\",\"answer\":\"It addresses how to allocate coolant distribution units across multiple parallel subloops and how to distribute flow fractions among them at exascale scale, which has not been systematically handled.\"},{\"question\":\"What does the three-layer optimization framework jointly determine?\",\"answer\":\"It jointly determines the integer partitioning of CDUs across subloops, continuous flow-fraction allocation, and per-timestep co-design of total flow rate and supply temperature under thermal safety constraints.\"},{\"question\":\"How is the digital twin and evaluation carried out in the proposed approach?\",\"answer\":\"A Modelica simulation model is built from Frontier data, and a reduced-order surrogate model evaluates 611 feasible CDU partitions across a full-year dataset of 49,353 timesteps.\"}]",1784175202,50,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"co-design-optimization-for-data-center-cooling-system-via-digital-twin","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/co-design-optimization-for-data-center-cooling-system-via-digital-twin/81655/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-25","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the paper address for exascale data center cooling plants?","Question",{"text":76,"@type":77},"It addresses how to allocate coolant distribution units across multiple parallel subloops and how to distribute flow fractions among them at exascale scale, which has not been systematically handled.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What does the three-layer optimization framework jointly determine?",{"text":81,"@type":77},"It jointly determines the integer partitioning of CDUs across subloops, continuous flow-fraction allocation, and per-timestep co-design of total flow rate and supply temperature under thermal safety constraints.",{"name":83,"@type":74,"acceptedAnswer":84},"How is the digital twin and evaluation carried out in the proposed approach?",{"text":85,"@type":77},"A Modelica simulation model is built from Frontier data, and a reduced-order surrogate model evaluates 611 feasible CDU partitions across a full-year dataset of 49,353 timesteps.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,114,119,122,126,129,133],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":29,"slug":113},"Technology","technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":22,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":127,"show_sort_weight":22,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":46,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":46,"category_name":135,"show_sort_weight":107,"slug":136},19,"General","general"]