{"id":{"repo_id":"ku","oai_identifier":"oai:kuscholarworks.ku.edu:1808/39402"},"canonical_url":"https://search.dev.ndltd.org/etd/ku/oai:kuscholarworks.ku.edu:1808/39402","repository":{"repo_id":"ku","name":"University of Kansas","base_url":"https://kuscholarworks.ku.edu/server/oai/request"},"display":{"title":"A Dynamic Resource Management Framework and Reconfiguration Strategies for Cloud-native Bulk Synchronous Parallel Applications","abstract":"Many High Performance Computing (HPC) applications following the Bulk Synchronous Parallel(BSP) model are increasingly deployed in cloud-native, multi-tenant container environments such as Kubernetes. Unlike dedicated HPC clusters, these shared platforms introduce resource virtualization and variability, making BSP applications more susceptible to performance fluctuations. Workload imbalance across supersteps can trigger the straggler effect, where faster tasks wait at synchronization barriers for slower ones, increasing overall execution time. Existing BSP resource management approaches typically assume static workloads and reuse a single configuration throughout execution. However, real-world workloads vary due to dynamic data and system conditions, making static configurations suboptimal. This limitation underscores the need for adaptive resource management strategies that respond to workload changes while considering reconfiguration costs. To address these limitations, we evaluate a dynamic, data-driven resource management framework tailored for cloud-native BSP applications. This approach integrates workload profiling, time-series forecasting, and predictive performance modeling to estimate task execution behavior under varying workload and resource conditions. The framework explicitly models the trade-off between performance gains achieved through reconfiguration and the associated checkpointing and migration costs incurred during container reallocation. Multiple reconfiguration strategies are evaluated, spanning simple window-based heuristics, dynamic programming methods, and reinforcement learning approaches. Through extensive experimental evaluation, this framework demonstrates up to 24.5% improvement in total execution time compared to a baseline static configuration. Furthermore, we systematically analyze the performance of each strategy under varying workload characteristics, simulation lengths, and checkpoint penalties, and provide guidance on selecting the most appropriate strategy for a given workload environment.","abstract_html":"Many High Performance Computing (HPC) applications following the Bulk Synchronous Parallel(BSP) model are increasingly deployed in cloud-native, multi-tenant container environments such as Kubernetes. Unlike dedicated HPC clusters, these shared platforms introduce resource virtualization and variability, making BSP applications more susceptible to performance fluctuations. Workload imbalance across supersteps can trigger the straggler effect, where faster tasks wait at synchronization barriers for slower ones, increasing overall execution time. Existing BSP resource management approaches typically assume static workloads and reuse a single configuration throughout execution. However, real-world workloads vary due to dynamic data and system conditions, making static configurations suboptimal. This limitation underscores the need for adaptive resource management strategies that respond to workload changes while considering reconfiguration costs. To address these limitations, we evaluate a dynamic, data-driven resource management framework tailored for cloud-native BSP applications. This approach integrates workload profiling, time-series forecasting, and predictive performance modeling to estimate task execution behavior under varying workload and resource conditions. The framework explicitly models the trade-off between performance gains achieved through reconfiguration and the associated checkpointing and migration costs incurred during container reallocation. Multiple reconfiguration strategies are evaluated, spanning simple window-based heuristics, dynamic programming methods, and reinforcement learning approaches. Through extensive experimental evaluation, this framework demonstrates up to 24.5% improvement in total execution time compared to a baseline static configuration. Furthermore, we systematically analyze the performance of each strategy under varying workload characteristics, simulation lengths, and checkpoint penalties, and provide guidance on selecting the most appropriate strategy for a given workload environment.","abstract_has_math":false,"creators":["Chitta, Krishna Chaitanya Reddy"],"institution":"University of Kansas","degree_name":"M.S.","degree_level":null,"degree_discipline":"Electrical Engineering & Computer Science","degree_department":null,"school":null,"contributors":[],"advisors":["Sun, Hongyang"],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026-05-31","date_published":"2026-05-31","updated_at":"2026-07-24T02:47:03Z","subjects":["BSP","Bulk Synchronous Parallel","Reinforcement Learning"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["https://www.proquest.com/LegacyDocView/DISSNUM/32670005"],"render_values":[{"text":"https://www.proquest.com/LegacyDocView/DISSNUM/32670005","href":"https://www.proquest.com/LegacyDocView/DISSNUM/32670005","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/1808/39402","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Sun, Hongyang"]},{"key":"dc:creator","label":"Author","values":["Chitta, Krishna Chaitanya Reddy"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-07-13T21:14:08Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-07-13T21:14:08Z"]},{"key":"dc:date.issued","label":"Date","values":["2026-05-31"]},{"key":"dc:publisher","label":"Institution","values":["University of Kansas"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical Engineering & Computer Science"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["BSP","Bulk Synchronous Parallel","Reinforcement Learning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["https://www.proquest.com/LegacyDocView/DISSNUM/32670005"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1808/39402"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Many High Performance Computing (HPC) applications following the Bulk Synchronous Parallel(BSP) model are increasingly deployed in cloud-native, multi-tenant container environments such as Kubernetes. Unlike dedicated HPC clusters, these shared platforms introduce resource virtualization and variability, making BSP applications more susceptible to performance fluctuations. Workload imbalance across supersteps can trigger the straggler effect, where faster tasks wait at synchronization barriers for slower ones, increasing overall execution time. Existing BSP resource management approaches typically assume static workloads and reuse a single configuration throughout execution. However, real-world workloads vary due to dynamic data and system conditions, making static configurations suboptimal. This limitation underscores the need for adaptive resource management strategies that respond to workload changes while considering reconfiguration costs. To address these limitations, we evaluate a dynamic, data-driven resource management framework tailored for cloud-native BSP applications. This approach integrates workload profiling, time-series forecasting, and predictive performance modeling to estimate task execution behavior under varying workload and resource conditions. The framework explicitly models the trade-off between performance gains achieved through reconfiguration and the associated checkpointing and migration costs incurred during container reallocation. Multiple reconfiguration strategies are evaluated, spanning simple window-based heuristics, dynamic programming methods, and reinforcement learning approaches. Through extensive experimental evaluation, this framework demonstrates up to 24.5% improvement in total execution time compared to a baseline static configuration. Furthermore, we systematically analyze the performance of each strategy under varying workload characteristics, simulation lengths, and checkpoint penalties, and provide guidance on selecting the most appropriate strategy for a given workload environment."]},{"key":"dc:title","label":"Title","values":["A Dynamic Resource Management Framework and Reconfiguration Strategies for Cloud-native Bulk Synchronous Parallel Applications"]}]}],"canonical_facts":{"dc:contributor.advisor":["Sun, Hongyang"],"dc:creator":["Chitta, Krishna Chaitanya Reddy"],"dc:date.accessioned":["2026-07-13T21:14:08Z"],"dc:date.available":["2026-07-13T21:14:08Z"],"dc:date.issued":["2026-05-31"],"dc:description.abstract":["Many High Performance Computing (HPC) applications following the Bulk Synchronous Parallel(BSP) model are increasingly deployed in cloud-native, multi-tenant container environments such as Kubernetes. Unlike dedicated HPC clusters, these shared platforms introduce resource virtualization and variability, making BSP applications more susceptible to performance fluctuations. Workload imbalance across supersteps can trigger the straggler effect, where faster tasks wait at synchronization barriers for slower ones, increasing overall execution time. Existing BSP resource management approaches typically assume static workloads and reuse a single configuration throughout execution. However, real-world workloads vary due to dynamic data and system conditions, making static configurations suboptimal. This limitation underscores the need for adaptive resource management strategies that respond to workload changes while considering reconfiguration costs. To address these limitations, we evaluate a dynamic, data-driven resource management framework tailored for cloud-native BSP applications. This approach integrates workload profiling, time-series forecasting, and predictive performance modeling to estimate task execution behavior under varying workload and resource conditions. The framework explicitly models the trade-off between performance gains achieved through reconfiguration and the associated checkpointing and migration costs incurred during container reallocation. Multiple reconfiguration strategies are evaluated, spanning simple window-based heuristics, dynamic programming methods, and reinforcement learning approaches. Through extensive experimental evaluation, this framework demonstrates up to 24.5% improvement in total execution time compared to a baseline static configuration. Furthermore, we systematically analyze the performance of each strategy under varying workload characteristics, simulation lengths, and checkpoint penalties, and provide guidance on selecting the most appropriate strategy for a given workload environment."],"dc:identifier.other":["https://www.proquest.com/LegacyDocView/DISSNUM/32670005"],"dc:identifier.uri":["https://hdl.handle.net/1808/39402"],"dc:language.iso":["en"],"dc:publisher":["University of Kansas"],"dc:subject":["BSP","Bulk Synchronous Parallel","Reinforcement Learning"],"dc:title":["A Dynamic Resource Management Framework and Reconfiguration Strategies for Cloud-native Bulk Synchronous Parallel Applications"],"dc:type":["Thesis"],"thesis:degree_discipline":["Electrical Engineering & Computer Science"],"thesis:degree_name":["M.S."]},"updated_at":"2026-07-24T02:47:03Z"}