• 综合
  • 标题
  • 关键词
  • 摘要
  • 学者
  • 期刊-刊名
  • 期刊-ISSN
  • 会议名称
搜索

作者:

Fang, J. (Fang, J..) (学者:方娟) | Zhang, X. (Zhang, X..) | Liu, S. (Liu, S..)

收录:

Scopus PKU CSCD

摘要:

In heterogeneous multi-core architecture, a large number of shared last level cache (LLC) space is occupied by GPU applications, leading to the system performance degradation. This paper presented a LLC buffer management strategy based on heterogeneous multi-core, buffer-for-GPU (BFG). A buffer was added on the side of LLC to filtrate streaming requests of GPU. Cache insensitive GPU messages directly accessed to buffer instead of accessing to LLC. Then, for different characteristics of CPU applications and GPU applications, a reasonable replacement strategy was adopted to increase the cache hit rate. Resulst show that compared with the traditional heterogeneous multi-core structure, the LLC buffer management strategy improves the CPU's and the GPU's IPC by 2.48% and 9.30% on average, and improves the hit rate of CPU applications by 70.45% on average. BFG can effectively filter the GPU requests, vacate space to CPU applications, and improve system performance. © 2019, Editorial Department of Journal of Beijing University of Technology. All right reserved.

关键词:

Cache management; Heterogeneous multi-core; Last level cache (LLC); Replacement strategy

作者机构:

  • [ 1 ] [Fang, J.]Faculty of Information Technology, Beijing University of Technology, Beijing, 100124, China
  • [ 2 ] [Zhang, X.]Beijing Intelligent City Insititute, Beijing University of Technology, Beijing, 100124, China
  • [ 3 ] [Liu, S.]Faculty of Information Technology, Beijing University of Technology, Beijing, 100124, China

通讯作者信息:

电子邮件地址:

查看成果更多字段

相关关键词:

相关文章:

来源 :

Journal of Beijing University of Technology

ISSN: 0254-0037

年份: 2019

期: 5

卷: 45

页码: 421-427

被引次数:

WoS核心集被引频次: 0

SCOPUS被引频次: 2

ESI高被引论文在榜: 0 展开所有

万方被引频次:

中文被引频次:

近30日浏览量: 0

归属院系:

在线人数/总访问数:206/4605275
地址:北京工业大学图书馆(北京市朝阳区平乐园100号 邮编:100124) 联系我们:010-67392185
版权所有:北京工业大学图书馆 站点建设与维护:北京爱琴海乐之技术有限公司