Session <Full Schedule · Contributors · Organizations · Search ProgramWorkshop: RSDHA: Redefining Scalability for Diversely Heterogeneous ArchitecturesSession ChairsSeyong LeeOak Ridge National Laboratory (ORNL)Mehmet E BelviranliColorado School of MinesAli AkogluUniversity of ArizonaEvent TypeWorkshopTimeSunday, 12 November 20239am - 12:30pm Location505Tags Accelerators Edge Computing Heterogeneous Computing Registration Categories W Presentations9:00am - 9:40am Invited Talk: Scaling Computing for Concurrent Data Structures Using Near-Memory Processing ArchitecturesPresenterIris Bahar9:40am - 9:56am Value-Based Resource Management at SoC ScaleAuthor/PresentersSerhan GenerSahil HassanAli Akoglu9:56am - 10:01am NVMe-Backed GNN Training on GPU Leveraging a Paged UVM Memory SystemAuthor/PresentersBenjamin WagleyPak MarkthubBo WuMehmet Belviranli10:01am - 10:31am RSDHA – Morning Break10:31am - 10:47am FFTX-IRIS: Toward Performance Portability and Heterogeneity for SPIRAL Generated CodeAuthor/PresentersSanil RaoHet MankadMohammad Alaul Haque MonilJeffrey VetterFranz Franchetti10:47am - 11:03am CHARM-SYCL: New Unified Programming Environment for Multiple Accelerator TypesAuthor/PresentersNorihisa FujitaBeau JohnstonRyohei KobayashiKeita TeranishiSeyong LeeTaisuke BokuJeffrey Vetter11:03am - 11:19am Vertical Scaling of Variational Multiscale Modeling for Fluid Dynamics: Successes, Challenges, and OpportunitiesAuthor/PresenterChristopher Coley11:19am - 11:35am Accelerator Integration in a Tile-Based SoC: Lessons Learned with a Hardware Floating Point Compression EngineAuthor/PresentersXueyang LiuPatricia Gonzalez-GuerreroIvy (Bo) PengRonald MinnichMaya Gokhale11:35am - 11:40am Evaluating Primitives in Deep Neural Network Libraries: A Case Study with the Softmax FunctionsAuthor/PresentersZheming JinJeffrey Vetter11:40am - 12:30pm RSDHA – Panel DiscussionModeratorsSeyong LeeMehmet BelviranliAli AkogluPanelistsMohamed WahibIvy PengUlya KarpuzcuSameer Shende
