BEGIN:VCALENDAR VERSION:2.0 PRODID:Linklings LLC BEGIN:VTIMEZONE TZID:America/Chicago X-LIC-LOCATION:America/Chicago BEGIN:DAYLIGHT TZOFFSETFROM:-0600 TZOFFSETTO:-0500 TZNAME:CDT DTSTART:19700308T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:-0500 TZOFFSETTO:-0600 TZNAME:CST DTSTART:19701101T020000 RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260202T201319Z LOCATION:266 DTSTART;TZID=America/Chicago:20251116T090000 DTEND;TZID=America/Chicago:20251116T173000 UID:submissions.supercomputing.org_SC25_sess221@linklings.com SUMMARY:The 12th Annual International Workshop on Innovating the Network f or Data-Intensive Science (INDIS) DESCRIPTION:This workshop aims to enhance the computer networking research track within the SC and HPC communities, highlighting innovations in high -performance networking, networking testbeds, and integrated research infr astructure. A particular focus for INDIS 2025 is the emergence of intellig ent networking within advanced cyber-infrastructure, including the use of AI, automation, and in-network telemetry and edge services. Moreover, the workshop aims to foster dialogue among experimental demonstrators and deve lopers of production-quality services, while promoting the reproducibility and wide adoption of the showcased research. This workshop also brings to gether participants in SCinet’s Network Research Exhibitions (NRE) and Exp erimental Networks of the Future (XNet) teams to present papers on their l atest innovations, designs, and solutions, and to showcase the next genera tion of networking challenges and solutions for HPC.\n\nNetwork Replay and Consistency Across Testbeds\n\nShared network testbeds are critical for s ystems and networking research. However, their shared hardware can introdu ce variability—like increased jitter or loss—that may impact experiment fi delity or reproducibility.\n\nWe present Choir, the first 100 Gbps replay tool designed to run on co...\n\n\nAlexander Wolosewicz (Illinois Institut e of Technology), Vinod Yegneswaran and Ashish Gehani (SRI), and Nik Sulta na (Illinois Institute of Technology)\n---------------------\nIntegrated I nfrastructure Panel\n\nIn this Integrated Research Infrastructure panel we will explore the network innovations needed for the emerging Artificial I ntelligence, Super- and Quantum Computing facilities in the context of the mission infrastructures and American Science Cloud. Moreover we will expl ore several different testb...\n\n\nNik Sultana (Illinois Institute of Tec hnology), Paul Ruth (Renaissance Computing Institute), Paola Gosso (Univer sity of Amsterdam), Frank Wuerthwein (UCSD), Marcos Schwarz (RNP), and Cod y Rotermund (Energy Sciences Network)\n---------------------\nRapid Quantu m Network Simulation Design with a Path to Scalable Execution\n\nAs quantu m networking grows in importance, its study is of interest to an ever wide r community. Several simulation frameworks allow for testing such systems on commodity hardware, but can be difficult to work with and performance- limited due to their predominantly serial nature. The SeQUeNCe simul...\n \n\nAaron Welch, Joel Dawson, and Mariam Kiran (Oak Ridge National Laborat ory (ORNL))\n---------------------\nFrom Path-Aware to Application-Aware S ource Routing using Traffic Classes\n\nSince the advent of software define d networking (SDN), its architecture allows better network flexibility, ca pacity planning and improved performance, especially for traffic engineeri ng. Additionally, network operators are using in-band network telemetry (I NT) to build efficient programmable network...\n\n\nShashwitha Puttaswamy (George Washington University) and Mariam Kiran (Oak Ridge National Labora tory (ORNL))\n---------------------\neCounter: Inline Per-IP Network Monit oring at Millisecond Resolution via eBPF\n\nScientific data acquisition (S ciDAQ) systems are shifting from archive-based workflows to streaming para digms, where real-time, fine-grained network monitoring becomes essential. While P4-enabled devices offer per-packet in-band observability, they req uire specialized switches and routers. Host-side...\n\n\nXinxin Mei, Jie C hen, Amitoj Singh, Ilya Baldin, and David Lawrence (Jefferson Lab)\n------ ---------------\nFrom Edge to HPC: Investigating Cross-Facility Data Strea ming Architectures\n\nIn this paper, we investigate three cross-facility d ata streaming architectures, Direct Streaming (DTS), Proxied Streaming (PR S), and Managed Service Streaming (MSS). We examine their architectural va riations in dataflow paths and deployment feasibility, and detail their im plementation using the DS2...\n\n\nAnjus George, Michael Brim, Christopher Zimmer, David Rogers, Sarp Oral, and Zach Mayes (Oak Ridge National Labor atory (ORNL))\n---------------------\nSENSE in Practice: Quantifying the E nd-to-End Benefits of Intent-Based Bandwidth Reservation for Exascale Scie nce Workflows\n\nThe escalating demands of scientific collaborations neces sitate advanced networking for deterministic, secure, and orchestrated ser vices across multiple administrative domains. The Software-Defined Network for End-to-end Networked Science at the Exascale (SENSE) paradigm address es these needs throug...\n\n\nInder Monga (ESnet; Lawrence Berkeley Nation al Laboratory (LBNL)); Mazahir Hussain (Korea Institute of Science and Tec hnology Information, University of Science and Technology (UST)); Justas B alcas (ESnet; Lawrence Berkeley National Laboratory (LBNL)); Aashay Arora and Diego Davila (University of California, San Diego (UCSD)); Buseung Cho (Korea Institute of Science and Technology Information, University of Sc ience and Technology (UST)); Xi Yang (ESnet; Lawrence Berkeley National La boratory (LBNL)); and Cees De Laat (University of Amsterdam)\n------------ ---------\nImplementing Network-level QoS at HPC Datacenters to Enable Dis tributed Scientific Workflows\n\nHigh-performance computing (HPC) datacent ers must simultaneously support real-time data streams with sub-millisecon d latency and bulk transfers requiring sustained multi-gigabit throughput— demands that compete for the same network resources. End-to-end performanc e guarantees are therefore esse...\n\n\nAnna Giannakou, Jonathan Skone, Vi nay Sawal, Ronal Kumar, Stephen Simms, Nicholas Wright, and Lavanya Ramakr ishnan (Lawrence Berkeley National Laboratory (LBNL))\n------------------- --\nOptimizing Network Resilience Using Domain-Specific Hardware Accelera tor for Dynamic Programming\n\nDynamic programming accelerators (DPAs) ar e devices designed with an instruction set optimized for dynamic programmi ng (DP) operations. DP is fundamental to solving complex networking proble ms, particularly those involving fault tolerance and routing under dynamic conditions. This paper explores the...\n\n\nAli Mazloum, Sergio Elizalde, Samia Choueiri, and Elie Kfoury (University of South Carolina); Jose Gome z (Fort Lewis College); Ali AlSabeh (University of South Carolina Aiken); and Jorge Crichigno (University of South Carolina)\n---------------------\ nStreamHub: High-performance Managed SciStream as a Service\n\nScientific data streaming enables the real-time transfer, processing, and analysis of high-throughput experimental data, providing low-latency insights that ar e essential for adaptive experiment control. We present SciStream-as-a-Ser vice (StreamHub), a secure, high-performance, and scalable framewor...\n\n \nSeena Vazifedunn (University of Chicago); Flavio Castro (Argonne Nationa l Laboratory (ANL)); Rajkumar Kettimuthu (Argonne National Laboratory (ANL ), University of Chicago); and Ian Foster and Kyle Chard (University of Ch icago, Argonne National Laboratory (ANL))\n---------------------\nComplex Parsing for In-Network Acceleration of High-Energy Physics Experiments\n\n This paper describes a novel application and evaluation of programmable ne tworking in High-Energy Physics (HEP): a complete parser for the custom pa cket format used by Fermilab’s DUNE experiment. Notably, this parser is im plemented on a Tofino programmable network switch and evaluated on the F.. .\n\n\nBjoern Sagstad and Nishanth Shyamkumar (Illinois Institute of Techn ology); Sophia Chen (Hinsdale South High School); Wesley Robert Ketchum (F ermilab); Roland Sipos (CERN); James Kowalkowski (Fermilab); Michael Wang (Fermi National Laboratory); and Nik Sultana (Illinois Institute of Techno logy, SCinet)\n---------------------\nMaking the Self-Driving “Net” Work: Developing Production-Ready ML Models for Self-Driving Networks\n\nThe vis ion of self-driving networks (AIOps) hinges on the ability to develop prod uction-ready machine learning models—models that are not only performant b ut also generalizable, robust, and trustworthy. Yet, most ML artifacts in networking today remain underspecified, suffering from shortcut l...\n\n\n Arpit Gupta (University of California, Santa Barbara)\n------------------- --\nAfternoon Break - International Workshop on Innovating the Network for Data Intensive Science (INDIS)\n---------------------\nScaling LLM Traini ng Using RDMA over Converged Ethernet\n\nWe present a comprehensive benchm arking study that evaluates the scaling performance of RDMA over Converged Ethernet (RoCE) and compares it with Infiniband in the context of large-s cale LLM training workloads. While Infiniband is traditionally favored for its low-latency, high-bandwidth characterist...\n\n\nAlex Batlle Casellas (Qualcomm Europe, Inc.); Adrián Pérez Diéguez (Qualcomm Technologies, Inc .); Aleix Torres-Camps (Qualcomm Europe, Inc.); Harris Teague (Qualcomm Te chnologies, Inc.); and Arnau Padres and Jordi Ros-Giralt (Qualcomm Europe, Inc.)\n---------------------\nLearning to Schedule: A Supervised Learning Framework for Network-Aware Scheduling of Data-Intensive Workloads\n\nDis tributed cloud environments hosting data-intensive applications often expe rience slowdowns from network congestion, asymmetric bandwidth, and inter- node data shuffling. These factors are typically not captured by host-leve l metrics such as CPU or memory. Scheduling without considering them can c a...\n\n\nSankalpa Timilsina and Susmit Shannigrahi (Tennessee Technologic al University)\n---------------------\nLunch break (on your own)\n-------- -------------\nModular Architecture for High-Performance and Low Overhead Data Transfers\n\nHigh-performance applications necessitate rapid and depe ndable transfer of massive datasets across geographically dispersed locati ons. Traditional file transfer tools often suffer from resource underutili zation and instability due to fixed configurations or monolithic optimizat ion methods. We propos...\n\n\nRasman Mubtasim Swargo (Missouri University of Science and Technology); Engin Arslan (Meta Plaforms, Inc.); and Md Ar ifuzzaman (Missouri University of Science and Technology)\n--------------- ------\nThe 12th Annual International Workshop on Innovating the Network f or Data Intensive Science (INDIS)\n\nThis workshop aims to enhance the com puter networking research track within the SC and HPC communities, highlig hting innovations in high-performance networking, networking testbeds, and integrated research infrastructure. A particular focus for INDIS 2025 is the emergence of intelligent networking w...\n\n\nEzra Kissel (Energy Scie nces Network (ESnet)); Rafael Guimaraes (Federal Institute of Espírito San to, Department of Informatics); Nik Sultana (Illinois Institute of Technol ogy); Cees De Laat (University of Amsterdam); Sonja Filiposka (Ss. Cyril a nd Methodius Univeristy in Skopje (UKIM) North Macedonia); and Akbar Kara (Ciena Corporation)\n---------------------\nTo Stream or Not to Stream: To wards A Quantitative Model for Remote HPC Processing Decisions\n\nModern s cientific instruments generate data at rates that increasingly outpace loc al compute capabilities, making traditional file-based workflows inadequat e for time-sensitive analysis and experimental steering. Real-time streami ng frameworks promise lower latency and improved efficiency, but lack ...\ n\n\nFlavio Castro, Weijian Zheng, Joaquin Chung, Ian Foster, and Raj Kett imuthu (Argonne National Laboratory (ANL))\n---------------------\nThe que st for high capacity, long reach, industrial strength networks\n\nRodney W ilson (Ciena)\n---------------------\nLLM-Based Optimization Algorithm Sel ection for High-Performance Networks Orchestration\n\nThe rapid growth of Artificial Intelligence (AI) applications, particularly through the widesp read adoption of Large Language Models (LLMs), has caused an unprecedented growth in computing and network infrastructures. Current infrastructure e xpansion cannot keep pace, resulting in suboptimal perform...\n\n\nAnestis Dalgkitsis, Cyril Hsu, Chrysa Papagianni, and Paola Grosso (University of Amsterdam) and Cees de Laat (University of Amsterdam, Lawrence Berkeley N ational Lab)\n---------------------\nMorning Break - International Worksho p on Innovating the Network for Data Intensive Science (INDIS)\n\nRecordin g: Livestreamed, Recorded\n\nRegistration Category: Technical Program Reg Pass, Workshop Reg Pass\n\nSession Chairs: Ezra Kissel (Energy Sciences Ne twork (ESnet)); Rafael Silva-Guimaraes (Federal Institute of Espírito Sant o, Department of Informatics); Nik Sultana (Illinois Institute of Technolo gy, SCinet); Cees de Laat (University of Amsterdam, Lawrence Berkeley Nati onal Laboratory (LBNL)); Sonja Filiposka (Ss. Cyril and Methodius Univeris ty in Skopje (UKIM) North Macedonia); and Akbar Kara (Ciena Corporation, S Cinet) END:VEVENT END:VCALENDAR