Optimization Techniques for GPU Programming
نویسندگان
چکیده
In the past decade, Graphics Processing Units have played an important role in field of high-performance computing and they still advance new fields such as IoT, autonomous vehicles, exascale computing. It is therefore to understand how extract performance from these processors, something that not trivial. This survey discusses various optimization techniques found 450 articles published last 14 years. We analyze optimizations different perspectives which shows are highly interrelated, explaining need for auto-tuning.
منابع مشابه
Programming Abstractions and Optimization Techniques for GPU-based Heterogeneous Systems
CPU/GPU heterogeneous systems have shown remarkable advantages in performance and energy consumption compared to homogeneous ones such as standard multi-core systems. Such heterogeneity represents one of the most promising trends for the near-future evolution of high performance computing hardware. However, as a double-edged sword, the heterogeneity also brings significant programming complexit...
متن کاملOptimization Techniques for CUDA Application Project Report of CIS6930 GPU: Parallel Architecture and Programming
In this paper, we summarize our experiment results of applying various optimization techniques for CUDA application running on NVIDIA Fermi GPUs. Our experiments on matrix multiplication and breadth first search algorithms show that optimization techniques such as coalesced global memory access, conflict-free shared memory access and data pre-fetching improve the performance of applications run...
متن کاملOCTAVIS: Optimization Techniques for Multi-GPU Multi-View Rendering
We present a high performance—yet low cost— system for multi-view rendering in virtual reality (VR) applications. In contrast to complex CAVE installations, which are typically driven by one render client per view, we arrange eight displays in an octagon around the viewer to provide a full 360◦ projection, and we drive these eight displays by a single PC equipped with multiple graphics units (G...
متن کاملOptimization Techniques for Mapping Algorithms and Applications onto CUDA GPU Platforms and CPU-GPU Heterogeneous Platforms
Title of dissertation: OPTIMIZATION TECHNIQUES FOR MAPPING ALGORITHMS AND APPLICATIONS ONTO CUDA GPU PLATFORMS AND CPU-GPU HETEROGENEOUS PLATFORMS Jing Wu, Doctor of Philosophy, 2014 Dissertation directed by: Professor Joseph F JaJa, Department of Electrical and Computer Engineering An emerging trend in processor architecture seems to indicate the doubling of the number of cores per chip every ...
متن کاملHigh-Level GPU Programming: Domain-Specific Optimization and Inference
When writing computer software one is often forced to balance the need for high runtime performance with high programmer productivity. By using a high-level language it is often possible to cut development times, but this typically comes at the cost of reduced run-time performance. Using a lower-level language, programs can be made very efficient but at the cost of increased development time. R...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: ACM Computing Surveys
سال: 2023
ISSN: ['0360-0300', '1557-7341']
DOI: https://doi.org/10.1145/3570638