.png){kind=icon}
.png){kind=icon}
Return
Improving GPU Memory Performance with Artificial Barrier Synchronization
.png){kind=icon}
PRE
AI
DOI：10.1109/TPDS.2013.133.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Barrier synchronization, an essential mechanism for a block of threads to guard data consistency, is regarded as a threat to performance. This study, however, provides a different viewpoint for barrier synchronization on GPUs: adding barrier synchronization, even when functionally unnecessary, can improve the performance of some memory-intensive applications. We explain this phenomenon using a memory contention model in which artificial barrier synchronization helps reduce memory contention and preserve data access locality. To yield practical applications, we identify a program pattern: artificial barrier synchronization can be used to synchronize the memory accesses when the data locality among threads is violated. Empirical results from three real-world applications demonstrate that artificial barrier synchronization can increase performance by 10 to 20 percent.
Keywords:
Graphics processors
synchronization
parallel languages
resource contention
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6
Papers:
5.2K
Citations:
1.1W