Oblivm: A programming framework for secure computation

C Liu, XS Wang, K Nayak, Y Huang… - 2015 IEEE Symposium …, 2015 - ieeexplore.ieee.org
We design and develop ObliVM, a programming framework for secure computation. ObliVM
offers a domain specific language designed for compilation of programs into efficient …

MGPUSim: Enabling multi-GPU performance modeling and optimization

Y Sun, T Baruah, SA Mojumder, S Dong… - Proceedings of the 46th …, 2019 - dl.acm.org
The rapidly growing popularity and scale of data-parallel workloads demand a
corresponding increase in raw computational power of Graphics Processing Units (GPUs) …

Beyond the socket: NUMA-aware GPUs

U Milic, O Villa, E Bolotin, A Arunkumar… - Proceedings of the 50th …, 2017 - dl.acm.org
GPUs achieve high throughput and power efficiency by employing many small single
instruction multiple thread (SIMT) cores. To minimize scheduling logic and performance …

The locality descriptor: A holistic cross-layer abstraction to express data locality in GPUs

N Vijaykumar, E Ebrahimi, K Hsieh… - 2018 ACM/IEEE 45th …, 2018 - ieeexplore.ieee.org
Exploiting data locality in GPUs is critical to making more efficient use of the existing caches
and the NUMA-based memory hierarchy expected in future GPUs. While modern GPU …

vSoC: Efficient Virtual System-on-Chip on Heterogeneous Hardware

J Qiu, Z Zhou, Y Li, Z Li, F Qian, H Lin, D Gao… - Proceedings of the …, 2024 - dl.acm.org
Emerging mobile apps such as UHD video and AR/VR access diverse high-throughput
hardware devices, eg, video codecs, cameras, and image processors. However, today's …

Griffin: Hardware-software support for efficient page migration in multi-gpu systems

T Baruah, Y Sun, AT Dinçer… - … Symposium on High …, 2020 - ieeexplore.ieee.org
As transistor scaling becomes increasingly more difficult to achieve, scaling the core count
on a single GPU chip has also become extremely challenging. As the volume of data to …

EAIS: Energy-aware adaptive scheduling for CNN inference on high-performance GPUs

C Yao, W Liu, W Tang, S Hu - Future Generation Computer Systems, 2022 - Elsevier
Recently, a large number of convolutional neural network (CNN) inference services have
emerged on high-performance Graphic Processing Units (GPUs). However, GPUs are high …

DRAGON: breaking GPU memory capacity limits with direct NVM access

P Markthub, ME Belviranli, S Lee… - … Conference for High …, 2018 - ieeexplore.ieee.org
Heterogeneous computing with accelerators is growing in importance in high performance
computing (HPC). Recently, application datasets have expanded beyond the memory …

Coda: Enabling co-location of computation and data for multiple gpu systems

H Kim, R Hadidi, L Nai, H Kim, N Jayasena… - ACM Transactions on …, 2018 - dl.acm.org
To exploit parallelism and scalability of multiple GPUs in a system, it is critical to place
compute and data together. However, two key techniques that have been used to hide …

Exploiting adaptive data compression to improve performance and energy-efficiency of compute workloads in multi-GPU systems

MK Tavana, Y Sun, NB Agostini… - 2019 IEEE International …, 2019 - ieeexplore.ieee.org
Graphics Processing Unit (GPU) performance has relied heavily on our ability to scale of
number of transistors on chip, in order to satisfy the ever-increasing demands for more …