01955nam 2200349 450 99657511640331620230420231959.01-66545-410-5(CKB)5590000001028476(NjHacI)995590000001028476(EXLCZ)99559000000102847620230420d2022 uy 0engur|||||||||||txtrdacontentcrdamediacrrdacarrier2022 IEEE/ACM Parallel Applications Workshop Alternatives To MPI+X (PAW-ATM) /IEEE Computer Society[Place of publication not identified] :IEEE Computer Society,2022.1 online resource1-66545-411-3 Title Page i pp. 1-1 -- Copyright Page pp. 2-2 -- Table of Contents pp. 3-4 -- Message from the Workshop Chairspp. 5-5 -- Workshop Organizationpp. 6-7 -- Extending OpenMP and OpenSHMEM for Efficient Heterogeneous Computingpp. 1-12 -- by Wenbin Lu, Shilei Tian, Tony Curtis, Barbara Chapman -- Task Fusion in Distributed Runtimespp. 13-25 -- by Shiv Sundram, Wonchan Lee, Alex Aiken -- Composition of Algorithmic Building Blocks in Template Task Graphspp. 26-38 -- by Thomas Herault, Joseph Schuchart, Edward F. Valeev, George Bosilca -- Asynchronous Workload Balancing through Persistent Work-Stealing and Offloading for a Distributed Actor Model Librarypp. 39-51 -- by Yakup Budanaz, Mario Wille, Michael Bader -- Design and Performance Evaluation of UCX for Tofu-D Interconnect with OpenSHMEM-UCX on Fugakupp. 52-61 -- by Yutaka Watanabe, Mitsuhisa Sato, Miwako Tsuji, Hitoshi Murai, Taisuke Boku.2022 IEEE/ACM Parallel Applications WorkshopHigh performance computingCongressesHigh performance computing004.11NjHacINjHaclBOOK9965751164033162022 IEEE2807321UNISA