Export Citations
Save this search
Please login to be able to save your searches and receive alerts for new content matching your search criteria.
- research-articleFebruary 2015
A comparative investigation of device-specific mechanisms for exploiting HPC accelerators
GPGPU-8: Proceedings of the 8th Workshop on General Purpose Processing using GPUsPages 1–12https://doi.org/10.1145/2716282.2716293A variety of computational accelerators have been greatly improved in recent years. Intel's MIC (Many Integrated Core) and both GPU architectures, NVIDIA's Kepler and AMD's Graphics Core Next, all represent real innovations in the field of HPC. Based ...
- research-articleFebruary 2015
Efficient utilization of GPGPU cache hierarchy
GPGPU-8: Proceedings of the 8th Workshop on General Purpose Processing using GPUsPages 36–47https://doi.org/10.1145/2716282.2716291Recent GPUs are equipped with general-purpose L1 and L2 caches in an attempt to reduce memory bandwidth demand and improve the performance of some irregular GPGPU applications. However, due to the massive multithreading, GPGPU caches suffer from severe ...
- research-articleFebruary 2015
GPU-SM: shared memory multi-GPU programming
GPGPU-8: Proceedings of the 8th Workshop on General Purpose Processing using GPUsPages 13–24https://doi.org/10.1145/2716282.2716286Discrete GPUs in modern multi-GPU systems can transparently access each other's memories through the PCIe interconnect. Future systems will improve this capability by including better GPU interconnects such as NVLink. However, remote memory access ...
- research-articleFebruary 2015
Adaptive GPU cache bypassing
GPGPU-8: Proceedings of the 8th Workshop on General Purpose Processing using GPUsPages 25–35https://doi.org/10.1145/2716282.2716283Modern graphics processing units (GPUs) include hardware- controlled caches to reduce bandwidth requirements and energy consumption. However, current GPU cache hierarchies are inefficient for general purpose GPU (GPGPU) comput- ing. GPGPU workloads tend ...
