Table of Contents

• Kernel-Initiated One-Sided Networking for GPU-Accelerated AI WorkloadsKhaled Hamidouche, John Bachan, Pak Markthub, Peter-Jan Gootzen, Elena Agostini, Sylvain Jeaugey, Aamir Shafi, Georgios Theodorakis, Manjunath Gorentla Venkata. 1-10 [doi]
• Mammoth: Macro-Level MPI Offloading to Off-Path Accelerator in DPUJong-Bin Lee, Pu-Rum Seo, Ki-Moon Jeong, Hyun-Wook Jin. 11-21 [doi]
• Scalable, Topology- and Multi-HCA-Aware Hierarchical GPU Allgather Using Parallel RingsAmirreza Barati Sedeh, Ryan Grant, Ahmad Afsahi. 22-32 [doi]
• Characterizing LLM Inference Energy-Performance Tradeoffs Across Workloads and GPU ScalingPaul Joe Maliakel, Shashikant Ilager, Ivona Brandic. 33-43 [doi]
• Quicktopia: Iteration-Level GPU Frequency Control for Energy-Latency Co-Optimization in LLM InferenceSoyang Baek, Bodon Jeong, Hongsu Byun, Sungyong Park. 44-53 [doi]
• ARKV: Adaptive and Resource-Efficient KV Cache Management Under Limited Memory Budget for Long-Context Inference in LLMsJianlong Lei, Shashikant Ilager. 54-63 [doi]
• ElastiFlow: Elastic Resource Management for Iterative Scientific Workflows in Hybrid HPC-Cloud InfrastructureSrishti Dasgupta, Kavitha Subramaniam, Michael Gerndt. 64-73 [doi]
• Efficiently Reproducing Distributed Workflows in Notebook-based SystemsTalha Azaz, Raza Ahmad, Md Saiful Islam, Douglas Thain, Tanu Malik. 74-83 [doi]
• Compass: Optimizing Compound AI Workflows for Dynamic AdaptationMilos Gravara, Juan Luis Herrera 0001, Stefan Nastic. 84-93 [doi]
• ADVICE: Automatic Identification of Variables to Checkpoint Through Compiler AugmentationXin Huang, Luanzheng Guo, Nathan R. Tallent, Kento Sato. 94-104 [doi]
• Enhanced SVM for Improving Application Performance Under GPU Memory OversubscriptionBennett Cooper, Thomas R. W. Scogland, Rong Ge 0002. 105-114 [doi]
• Priority-Aware GPU Co-Scheduling for High Performance ComputingNaman Kulshreshtha, Tapasya Patki, Aniruddha Marathe, Tom Scogland, Rong Ge 0002. 115-125 [doi]
• EmbdC: Error-Bounded Lossy Video Embedding Compression for On-Device LLM InferenceBo Jiang, Taolue Yang, Youyuan Liu, Xubin He, Sheng Di, Sian Jin. 126-135 [doi]
• LLM-Pilot: SLO-Aware and Cost-Efficient LLM Serving on Public Cloud VM Clusters via OffloadingJinwoo Kim, Kihyun Kim, Hyunsun Chung, Jihoon Yang, James J. Kim, Dong Li, Youngjae Kim. 136-146 [doi]
• EAFAL: An Edge-Based Agentic Framework for Adaptive Selection Between SLMs and LLMsChamara Madarasingha, Prajyot Singh, Redowan Mahmud, Mahbuba Afrin, Aneesh Krishna, Salil S. Kanhere. 147-156 [doi]
• Exploring Silent Data Corruption as a Reliability Challenge in LLM TrainingAnton Altenbernd, Philipp Wiesner, Odej Kao. 157-166 [doi]
• Noise-Aware Misclassification Attack Detection in Collaborative DNN InferenceShima Yousefi, Saptarshi Debroy. 167-177 [doi]
• SeCI: A Framework for Self-Certified Identity for Autonomous AI AgentsNehal F. Al-Otaiby, Mohammad Hammoudeh, Jameleddine Hassine. 178-187 [doi]
• GASched: Goal-Adaptive Hierarchical Reinforcement Learning for Multi-Objective HPC Job SchedulingMinsol Choo, Sangyoon Oh 0001. 188-197 [doi]
• Reducing Backfill Failures From Workload Drift with Lightweight Uncertainty Buffers in HPC Job SchedulingJiheon Choi, Sangyoon Oh 0001. 198-209 [doi]
• Automated Configuration of Power-Management Knobs for Optimal HPC Job ExecutionsFrancesco Antici, Andrea Proia, Ryoma Ohara, Toshihiro Hanawa, Zeynep Kiziltan, Andrea Bartolini, Jens Domke. 210-220 [doi]
• Reuse-Aware Min-Cost Flow Scheduling for Distributed LLM Training in Hybrid Optical-Electrical NetworksYifan Yang, Gongming Zhao, Hongli Xu 0001, Huihui Tang. 221-230 [doi]
• Tula: Optimizing Time, Cost, and Generalization in Distributed Large-Batch TrainingSahil Tyagi, Feiyi Wang. 231-242 [doi]
• VeX: Scaling HNSW-Based Vector Search with DPU Memory and ParallelismKihwan Kim, Hyungsun Yoo, Woojung Kim, Donghyun Min, Myungcheol Lee, Jihoon Yang, Weikuan Yu, Youngjae Kim 0001. 243-253 [doi]
• Green or Fast? Learning to Balance Cold Starts and Idle Carbon in Serverless ComputingBowen Sun, Christos D. Antonopoulos, Evgenia Smirni, Bin Ren 0002, Nikolaos Bellas, Spyros Lalis. 254-264 [doi]
• OpenDC-STEAM: Realistic Modeling and Systematic Exploration of Composable Techniques for Sustainable DatacentersDante Niewenhuis, Sacheendra Talluri, Alexandru Iosup, Tiziano De Matteis. 265-275 [doi]
• Spanergy: Energy-Aware Distributed Tracing for MicroservicesCésar Perdigão Batista, Denis Conan, Sophie Chabridon. 276-286 [doi]
• Larger Cloud Servers, Fewer Hosts? on the Evolution of VM Sizes in IaaS PlatformsPierre Jacquet, Camille Coti, Marcos Dias de Assunção. 287-296 [doi]
• TelePod: Live Migration for Stateful ContainersMingjie Yan, Atharva Ranade, Xin Zhang, Kartik Gopalan. 297-306 [doi]
• Cost-Justified Multi-type Resource Fair Scheduling for KubernetesBo Yan, Madhusudhan Govindaraju, Sujoy Sikdar. 307-312 [doi]
• Xfagent: Automating Multi-Cloud Deployment of Agentic Workflows on Faas PlatformsVarad Kulkarni, Vaibhav Jha, Nikhil Reddy, Anand Eswaran, Praveen Jayachandran, Yogesh Simmhan. 313-318 [doi]
• A Multi-Armed Bandit-Based Participant Selection Method for Federated Recommendation SystemsJintao Liu, Mohammad Goudarzi, Adel Nadjaran Toosi. 319-328 [doi]
• FedMO: Mobility-Aware Client Selection in Federated Learning for Drone Delivery SystemsYuan Jiang, Xiao Liu 0004, Jia Xu 0010, Aiting Yao, Frank Jiang 0001, Xuejun Li 0001. 329-338 [doi]
• Heterogeneous-Resource-Aware Federated Learning with Intelligent LoRA Allocation and AggregationYouye Xie, Yao Lian, Kevin Chen, Abdul-Latif, Lingzhi Zhao, Reza Farivar. 339-349 [doi]
• AdaSched: A Performance-Driven Cluster Scheduler for Deep Learning Workloads using Deep Reinforcement LearningHan Yin, Jialun Li, Xuan Mo, Weigang Wu. 350-359 [doi]
• Congestion-Aware Pricing for Fast and Efficient Edge-Cloud ComputingPolyzois Soumplis, Emmanouel Varvarigos. 360-369 [doi]
• SD-MoE: Scenario-Driven MoE Forecasting for Intelligent Elastic Scaling in Cloud ClustersXianzhao Guo, Weipeng Cao, Minxian Xu, Dachuan Li, Chuanfei Xu, Xi Tao, Zhong Ming 0001. 370-379 [doi]
• Orchestrating WASM-Based MCP Tool Runtimes for AI Agents Across Edge-Cloud ContinuumMoohyun Song, Hayoung Kim, Kyoohyun Lee, Jae Gi Son, Kyungyong Lee 0001. 380-391 [doi]
• SwiftBot: A Decentralized Platform for LLM-Powered Federated Robotic Task ExecutionYueming Zhang, Shuai Xu, Zhengxiong Li, Fangtian Zhong, Xiaokun Yang, Hailu Xu. 392-401 [doi]
• DriveCache: On-Board Compute Caching for Scalable Vehicular Edge Computing NetworksSuvarthi Sarkar, Aditya Gupta, Salil Kashyap, Aryabartta Sahu. 402-411 [doi]
• Dual-Distilled Heterogeneous Federated Learning with Adaptive Margins for Trainable Global PrototypesFatema Siddika, Md. Anwar Hossen, Wensheng Zhang, Anuj Sharma 0001, J. Pablo Muñoz, Ali Jannesari. 412-421 [doi]
• RDA-CAFL: Reputation-Dynamic and Distillation-based Asynchronous Conflict-Aware Federated Learning for UAV NetworksJie Li, Pengyang Li, Liming Sun, Xingwei Wang, Yihang Zhang. 422-430 [doi]
• Evaluating Federated Learning Beyond Simulation: A Deployment-Aware MethodologyMathis Valli, Cédric Tedeschi, Gabriel Antoniu, Loïc Cudennec, Alexandru Costan. 431-441 [doi]
• Making Variable-Size I/O Practical in ZNS SSDsSijie Lan, Abutalib Aghayev, Mahmut T. Kandemir, Umesh Maheshwari. 442-451 [doi]
• NIO-Cache: Device-Affinitive Page Cache Placement Mechanism for NUMA SystemsJiazheng Zhang, Xiaoyang Wang 0006, Jiwu Shu. 452-461 [doi]
• S-MSHR: A Scalable MSHR Architecture Using Cache Tag Data-Ready Bits and Index QueuesShuang Wu, Xu Zhang 0033, Yibin Xu, Yangyang Zhao, Tianyue Lu, Mingyu Chen 0001. 462-471 [doi]
• Benchmarking Message Brokers for IoT Edge Computing: A Comprehensive Performance StudyTapajit Chandra Paul, Pawissanutt Lertpongrujikorn, Hai Duc Nguyen 0005, Mohsen Amini Salehi. 472-481 [doi]
• Adaptive Edge Context Caching Framework for Context-Aware Internet of Things ApplicationsAshish Manchanda, Prem Prakash Jayaraman, Abhik Banerjee, Arkady B. Zaslavsky. 482-487 [doi]
• Hierarchical Semi-Supervised Federated Learning for UAV-Enabled Fire MonitoringMark Adrian Gambito, Bahman Javadi, Lorenzo Carnevale, Massimo Villari. 488-493 [doi]
• Synthetic Data Generation for Storage Failure Prediction in Large-Scale SystemsChandranil Chakraborttii, Ana Luisa Veroneze Solórzano, Devesh Tiwari. 494-499 [doi]
• FedOort: A Fair and Efficient Optimization Method for Federated LearningZhaohua Zheng, Junhui Du, Xin Xie, Qiquan Chen, Xu Han, Qijun Huang, Jin Zhang. 500-509 [doi]
• Evidential Trust-Aware Model Personalization in Decentralized Federated Learning for Wearable IoTMurtaza Rangwala, Richard O. Sinnott, Rajkumar Buyya. 510-519 [doi]
• CroSatFL: Energy-Efficient Federated Learning with Cross-Aggregation for Satellite Edge ComputingNan Yang, Bahman Javadi, Rodrigo Neves Calheiros, David Boland, Philip Leong. 520-529 [doi]
• ARC: Adaptive Resource Coordination for Write Stall Mitigation in LSM-TreeGuangxun Zhao, Yongjie Zhu, Suhwan Shin, Seehwan Yoo, Jongmoo Choi. 530-539 [doi]
• ALTOCUMULUS: Enabling Efficient Erasure Coding in IPFSMohammad Rizk, Shadi Ibrahim, Thomas Lambert. 540-549 [doi]
• BucketLSM: Breaking the Compaction Scalability Barrier in LSM-Based Key-Value StoresJaewan Park, KyungWook Min, Sungjin Byeon, Taewan Noh, Hyungi Park, Xubin He, Hong-Yeon Kim, Youngjae Kim 0001. 550-560 [doi]
• Multi-Objective Load Balancing for Heterogeneous Edge-Based Object Detection SystemsDaghash K. Alqahtani, Maria A. Rodriguez 0001, Muhammad Aamir Cheema, Adel Nadjaran Toosi. 561-570 [doi]
• Deep Reinforcement Learning-Driven Edge Offloading for Latency-Constrained XR PipelinesSourya Saha, Saptarshi Debroy. 571-582 [doi]
• Decentralized Resource Sharing in Edge-Cloud Federations via Multi-Agent Hierarchical Reinforcement LearningPanagiotis Kokkinakis, Polyzois Soumplis, Emmanouel Varvarigos. 583-592 [doi]
• CloudFormer: An Attention-Based Performance Prediction for Public Clouds with Unknown WorkloadAmirhossein Shahbazinia, Darong Huang 0003, Luis Costero, David Atienza 0001. 593-603 [doi]
• PM2Lat: Highly Accurate and Generalized Prediction of DNN Execution Latency on GPUsTruong Thanh Le, Hoang-Loc La, Amir Taherkordi, Frank Eliassen, Phuong Hoai Ha, Peiyuan Guan. 604-614 [doi]
• Towards Proactive AIOps: Transfer Learning for Unsupervised Anomaly Detection via Bi-LSTM AutoencoderDanny De Novi, Lorenzo Carnevale, Massimo Villari. 615-624 [doi]
• Efficient Concurrent GHZ State Distribution in Quantum NetworksCharles Cao 0001, Weisheng Si, Jong Youl Choi, Sajal K. Das 0001. 625-634 [doi]
• QRAP: Adaptive Quantum-Safe Risk-Aware Prioritization for Cloud ApplicationsSurabhi Garg, Nidhi Singh, Delton Myalil, Shiv Shankar, Meena Singh Dilip Thakur, Rajan Ma. 635-640 [doi]
• BCRLSecureLink: A Blockchain, Cryptography, and Reinforcement Learning-Based Defense Against Link Discovery Attacks in SDVNPatikiri Arachchige Don Shehan Nilmantha Wijesekara, Harsha S. Gardiyawasam Pussewalage, Kalupahana Liyanage Kushan Sudheera, Geeth Priyankara Wijesiri. 641-646 [doi]
• Lifting to Tensors when Compiling Scientific Computing Workloads for AI EnginesNick Brown 0002, Gabriel Rodriguez-Canal. 647-652 [doi]
• Kumo: A Security-Focused Serverless Cloud SimulatorWei Shao, Khaled N. Khasawneh, Setareh Rafatirad, Houman Homayoun, Chongzhou Fang. 653-664 [doi]
• LogLearners: Identifying Compromised AI Functions in Serverless SystemsAdil Bin Bhutto, Erik Elmroth, Monowar Bhuyan. 655-674 [doi]
• PoliFlow: Inferring Control-Flow Policies from Serverless WorkflowsPedro Escaleira, Vitor A. Cunha, João Paulo Barraca, Diogo Gomes 0001, Rui L. Aguiar. 675-685 [doi]
• ACNNS: A Multi-Interest Recommendation Model with Capsule NetworkYan Zhang, Xiaotong Cui, Nan Wang 0024, Yingli Zhong. 686-694 [doi]
• Inversetune: Inverse Synthetic Fine-Tuning for Reliable Structured Output in Small Language ModelsMarkus Götz, Hojjat Baghban. 695-700 [doi]
• SOtrain: Efficient LLM Fine-Tuning on a Single-GPU Commodity Server via User-Space I/O and Scheduling OptimizationsZhengguo Liu, Hao Lan, Jiwu Shu. 701-706 [doi]
• RapidGNN: Communication-Efficient Distributed Training on Large-Scale Graph Neural NetworksArefin Niam, Tevfik Kosar, Md. S. Q. Zulkar Nine. 707-712 [doi]