Supporting Very Large DRAM Caches with Compound-Access Scheduling and MissMap

Supporting Very Large DRAM Caches with Compound-Access Scheduling and MissMap
复制标题

通过复合访问调度和 MissMap 支持超大型 DRAM 缓存

DOI：
--
复制
发表时间：
2012
期刊：
IEEE Micro
影响因子：
3.6
通讯作者：
M. Hill
M. Hill
中科院分区：
计算机科学3区
文献类型：
--
作者：
G. Loh;M. Hill

文献摘要

被引文献

相似文献

这项工作有效地使传统的块大小非常大的裸片堆叠DRAM高速缓存与两个创新：它使命中更快的复合访问调度和错过更快的MissMap。这些机制的结合使新的组织提供的性能与理想的DRAM缓存，采用不切实际的大基于SRAM的片上标签阵列。
This work efficiently enables conventional block sizes for very large die-stacked DRAM caches with two innovations: it makes hits faster with compound-access scheduling and misses faster with a MissMap. The combination of these mechanisms enables the new organization to deliver performance comparable to that of an idealistic DRAM cache that employs an impractically large SRAM-based on-chip tag array.