Automatically exploiting implicit Pipeline Parallelism from multiple dependent kernels for GPUs

No abstract is available for this paper.

• Publication year

  2016

• Venue

  International Conference on Parallel Architectures and Compilation Techniques

• Publication date

  2016-09-11

• Fields of study

  Computer Science, Engineering

• Identifiers
  DOI 10.1145/2967938.2967952
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Towards high performance paged memory for GPUs

  2016influential reference
• Automatic Parallelization of Kernels in Shared-Memory Multi-GPU Nodes

  2015cited by this paper
• GPUdrive: Reconsidering Storage Accesses for GPU Acceleration

  2014cited by this paper
• A paradigm shift in GP-GPU computing: task based execution of applications with dynamic data dependencies

  2014cited by this paper
• COMP: Compiler Optimizations for Manycore Processors

  2014cited by this paper
• An investigation of Unified Memory Access performance in CUDA

  2014cited by this paper
• GPUdmm: A high-performance and memory-oblivious GPU architecture using dynamic memory management

  2014cited by this paper
• Improving GPGPU resource utilization through alternative thread block scheduling

  2014cited by this paper
• GPUnet

  2014cited by this paper
• Whippletree

  2014cited by this paper
• Reducing GPU offload latency via fine-grained CPU-GPU synchronization

  2013cited by this paper
• GPU Technology Conference

  2013cited by this paper
• GPUfs: Integrating a file system with GPUs

  2013cited by this paper
• The case for GPGPU spatial multitasking

  2012cited by this paper
• Auto-tuning a high-level language targeted to GPU codes

  2012cited by this paper
• Dynamically managed data for CPU-GPU architectures

  2012cited by this paper
• Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing

  2012cited by this paper
• Achieving a single compute device image in OpenCL for multiple GPUs

  2011cited by this paper
• Where is the data? Why you cannot debate CPU vs. GPU performance without the answer

  2011cited by this paper
• Dymaxion: Optimizing memory access patterns for heterogeneous systems

  2011cited by this paper
• An asymmetric distributed shared memory model for heterogeneous parallel systems

  2010cited by this paper
• Analyzing CUDA workloads using a detailed GPU simulator

  2009cited by this paper
• GRAMPS

  2009cited by this paper
• Rodinia: A benchmark suite for heterogeneous computing

  2009cited by this paper
• LLVM: a compilation framework for lifelong program analysis & transformation

  2004cited by this paper
• A method for overlapping and erasure of lists

  1960cited by this paper

• GSOHC: Global Synchronization Optimization in Heterogeneous Computing

  2025cites this paper
• BlockMaestro: Enabling Programmer-Transparent Task-based Execution in GPU Systems

  2021cites this paper
• Pipelined CPU-GPU Scheduling for Caches

  2021cites this paper
• RTGPU: Real-Time GPU Scheduling of Hard Deadline Parallel Tasks With Fine-Grain Utilization

  2021cites this paper
• Pipelined CPU-GPU Scheduling to Reduce Main Memory Accesses

  2021cites this paper
• A framework for scheduling dependent programs on GPU architectures

  2020cites this paper
• GOPipe: A Granularity-Oblivious Programming Framework for Pipelined Stencil Executions on GPU

  2020cites this paper
• SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs

  2019cites this paper
• A survey of architectural approaches for improving GPGPU performance, programmability and heterogeneity

  2019cites this paper
• HiWayLib: A Software Framework for Enabling High Performance Communications for Heterogeneous Pipeline Computations

  2019cites this paper
• A case study on machine learning for synthesizing benchmarks

  2019cites this paper
• Big Data Dwarfs: Towards Fully Understanding Big Data Analytics Workloads

  2018cites this paper
• Data motifs: a lens towards fully understanding big data and AI workloads

  2018cites this paper
• Scheduling Methods to Optimize Dependent Programs for GPU Architecture

  2018cites this paper
• VersaPipe: A Versatile Programming Framework for Pipelined Computing on GPU

  2017cites this paper