Dependency Aware GPGPU Kernel Scheduling
碩士 === 國立臺灣科技大學 === 資訊工程系 === 106 === Recent GPUs are widely used in a variety of areas, such as image processing, deep learning, artificial intelligence, and so on. Related researches are constantly being presented. And programs such as deep learning and artificial intelligence do not use only on...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2017
|
Online Access: | http://ndltd.ncl.edu.tw/handle/5xsqa9 |
id |
ndltd-TW-106NTUS5392003 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-106NTUS53920032019-05-15T23:46:36Z http://ndltd.ncl.edu.tw/handle/5xsqa9 Dependency Aware GPGPU Kernel Scheduling 改善相依性程式之排程法 Wei-Cheng Liao 廖偉丞 碩士 國立臺灣科技大學 資訊工程系 106 Recent GPUs are widely used in a variety of areas, such as image processing, deep learning, artificial intelligence, and so on. Related researches are constantly being presented. And programs such as deep learning and artificial intelligence do not use only one kernel to execute, but assign tasks to different sub-kernels, which are data dependency. We categorize them as dependent kernels. In the case of current GPU programming, these dependent kernels will be implemented in a sequential manner because of data dependencies, which will result in a reduction in GPU parallelism. In addition, the GPU will usually save the processed data back to memory, but for the dependent core, the data will be used again, resulting in unnecessary memory access. We proposed the method to break the rules that the dependency kernel must perform in sequence by modifying the scheduling method in the simulator and retains the data in the cache memory with the appropriate memory write back policy. Yuan-Shin Hwang 黃元欣 2017 學位論文 ; thesis 41 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺灣科技大學 === 資訊工程系 === 106 === Recent GPUs are widely used in a variety of areas, such as image processing, deep learning, artificial intelligence, and so on. Related researches are constantly being presented. And programs such as deep learning and artificial intelligence do not use only one kernel to execute, but assign tasks to different sub-kernels, which are data dependency. We categorize them as dependent kernels.
In the case of current GPU programming, these dependent kernels will be implemented in a sequential manner because of data dependencies, which will result in a reduction in GPU parallelism. In addition, the GPU will usually save the processed data back to memory, but for the dependent core, the data will be used again, resulting in unnecessary memory access.
We proposed the method to break the rules that the dependency kernel must perform in sequence by modifying the scheduling method in the simulator and retains the data in the cache memory with the appropriate memory write back policy.
|
author2 |
Yuan-Shin Hwang |
author_facet |
Yuan-Shin Hwang Wei-Cheng Liao 廖偉丞 |
author |
Wei-Cheng Liao 廖偉丞 |
spellingShingle |
Wei-Cheng Liao 廖偉丞 Dependency Aware GPGPU Kernel Scheduling |
author_sort |
Wei-Cheng Liao |
title |
Dependency Aware GPGPU Kernel Scheduling |
title_short |
Dependency Aware GPGPU Kernel Scheduling |
title_full |
Dependency Aware GPGPU Kernel Scheduling |
title_fullStr |
Dependency Aware GPGPU Kernel Scheduling |
title_full_unstemmed |
Dependency Aware GPGPU Kernel Scheduling |
title_sort |
dependency aware gpgpu kernel scheduling |
publishDate |
2017 |
url |
http://ndltd.ncl.edu.tw/handle/5xsqa9 |
work_keys_str_mv |
AT weichengliao dependencyawaregpgpukernelscheduling AT liàowěichéng dependencyawaregpgpukernelscheduling AT weichengliao gǎishànxiāngyīxìngchéngshìzhīpáichéngfǎ AT liàowěichéng gǎishànxiāngyīxìngchéngshìzhīpáichéngfǎ |
_version_ |
1719154153158082560 |