Visible to the public Addressing Global Data Dependencies in Heterogeneous Asynchronous Runtime Systems on GPUs

TitleAddressing Global Data Dependencies in Heterogeneous Asynchronous Runtime Systems on GPUs
Publication TypeConference Paper
Year of Publication2017
AuthorsPeterson, Brad, Humphrey, Alan, Schmidt, John, Berzins, Martin
Conference NameProceedings of the Third International Workshop on Extreme Scale Programming Models and Middleware
PublisherACM
Conference LocationNew York, NY, USA
ISBN Number978-1-4503-5133-1
KeywordsAsynchronous Many-Task, Coal Boiler, Collaboration, composability, Data dependencies, GPU, middleware security, policy, policy-based governance, Programming Models, pubcrawl, Radiative Heat Transfer, resilience, Resiliency, Runtime Systems, Scalability, Uintah
AbstractLarge-scale parallel applications with complex global data dependencies beyond those of reductions pose significant scalability challenges in an asynchronous runtime system. Internodal challenges include identifying the all-to-all communication of data dependencies among the nodes. Intranodal challenges include gathering together these data dependencies into usable data objects while avoiding data duplication. This paper addresses these challenges within the context of a large-scale, industrial coal boiler simulation using the Uintah asynchronous many-task runtime system on GPU architectures. We show significant reduction in time spent analyzing data dependencies through refinements in our dependency search algorithm. Multiple task graphs are used to eliminate subsequent analysis when task graphs change in predictable and repeatable ways. Using a combined data store and task scheduler redesign reduces data dependency duplication ensuring that problems fit within host and GPU memory. These modifications did not require any changes to application code or sweeping changes to the Uintah runtime system. We report results running on the DOE Titan system on 119K CPU cores and 7.5K GPUs simultaneously. Our solutions can be generalized to other task dependency problems with global dependencies among thousands of nodes which must be processed efficiently at large scale.
URLhttp://doi.acm.org/10.1145/3152041.3152082
DOI10.1145/3152041.3152082
Citation Keypeterson_addressing_2017