Nupar

Ukidave, Yash; Paravecino, Fanny Nina; Yu, Liang; Kalra, Charu; Momeni, Amir; Chen, Zhongliang; Materise, Nicholas; Daley, Brett; Mistry, Perhaad; Kaeli, David

doi:10.1145/2668930.2688046

Cited by 26 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1. Connected Component Labelling (CCL) is a well-known labeling algorithm that is commonly used for object detection [51].…”

Section: Benchmarksmentioning

confidence: 99%

Compiler and runtime based parallelization & optimization for GPUs

Ozen¹

View full text Add to dashboard Cite

Graphics Processing Units (GPU) have been widely adopted to accelerate the execution of HPC workloads due to their vast computational throughput, ability to execute a large number of threads inside SIMD groups in parallel and their use of hardware multithreading to hide long pipelining and memory access latencies. There are two APIs commonly used for native GPU programming: CUDA, which only targets NVIDIA GPUs and OpenCL, which targets all types of GPUs as well as other accelerators. However these APIs only expose low-level hardware characteristics to the programmer. So developing applications able to exploit the dazzling performance of GPUs is not a trivial task, and becomes even harder when they have irregular data access patterns or control flows. Several approaches have been proposed to help simplify accelerator programming. Models like OpenACC and OpenMP are intended to solve the aforementioned programming challenges. They take a directive based approach which allows the users to insert non-executable directives that guide the compiler to handle the low-level complexities of the system. However they have a performance gap with native programming models as their compiler does not have comprehensive knowledge about how to transform code and what to optimize. This thesis targets directive-based programming models to enhance their capabilities for GPU programming. The thesis introduces a new dialect model, which is a combination of OpenMP and OmpSs. It also includes several extensions and the MACC infrastructure, a source-to-source compiler targeting CUDA developed on top of BSC's Mercurium compiler and able to support the new dialect model. The new model allows the use of multiple GPUs in conjunction with the vector and heavily multithreaded capabilities in multicore processors automatically. Moreover, it introduces new clauses to make use of on-chip memory efficiently. Secondly the thesis focusses on code transformation techniques and proposes the LazyNP method to support nested parallelism for irregular applications such as sparse matrix operations, graph and graphics algorithms. The method efficiently increases thread granularity for the code region where nested parallelism is desired. The compiler generates code to dynamically pack kernel invocations and to postpone their execution until a bunch of them are available. To the best of our knowledge, LazyNP code transformation was the first successful code transformation method related to nested directives for GPUs. Finally, the thesis conducts a thorough exploration of conventional loop scheduling methods on GPUs to find the advantage and disadvantages of each method. It then proposes the concept of optimized dynamic loop scheduling as an improvement to all the existing methods. The contributions of this thesis improve the programmability of GPUs. This has had an outstanding impact on the whole OpenMP and OpenACC language committee. Additionally, our work includes contributions to widely used compilers such as Mercurium, Clang and PGI, helping thousands of users to take advantage of our work.

show abstract

“…1. Connected Component Labelling (CCL) is a well-known labeling algorithm that is commonly used for object detection [51].…”

Section: Benchmarksmentioning

confidence: 99%

Compiler and runtime based parallelization & optimization for GPUs

Ozen¹

View full text Add to dashboard Cite

show abstract

“…To date, GPU benchmarks fall into one of two categories. They either evaluate general-purpose GPU computing capabilities [35,36,44,146,170,181], or target assessment of the performance of a specific class of workloads [106,110,171] AIBench [63] is an industry-initiated benchmark suite that is focused on industrial AI services.…”

Section: Graph Computing Benchmark Suitesmentioning

confidence: 99%

Enabling accelerators for graph computing

Shivdikar

View full text Add to dashboard Cite

show abstract

“…GPUs were originally designed for processing vertexes and polygons, rendering 3D objects for video games, and displaying information on screens in near real-time. It was not until the introduction of programmable shaders and high-level languages that GPUs became a significant accelerator for a wide class of computations [60], especially in the scientific computing community.…”

Section: Gpu Architecture and Programming Frameworkmentioning

confidence: 99%

Towards real-time safe flight paths for urban air mobility

Gutierrez

View full text Add to dashboard Cite

Nupar

Cited by 26 publications

References 19 publications

Compiler and runtime based parallelization & optimization for GPUs

Compiler and runtime based parallelization & optimization for GPUs

Enabling accelerators for graph computing

Towards real-time safe flight paths for urban air mobility

Contact Info

Product

Resources

About