期刊文献+

一种基于活跃周期的低端口数低能耗寄存器堆设计 被引量:1

Active-Cycle Based Register File Design for Reduced Ports and Energy
下载PDF
导出
摘要 多端口寄存器堆有助于挖掘指令级和线程级并行性,但同时带来面积、能耗和访问时间的压力.文章面向超标量和SMT处理器,给出了一种方法,即通过增加一个小的活跃值堆(Active Value File,AVF)选择性地保存处于活跃周期(从产生到最后一次使用之间)的物理寄存器值.AVF结构可分担主寄存器堆的访问压力并降低端口数目,实现简单且具有写过滤的特点.在获得较大幅度能耗降低的同时不影响时钟频率且IPC损失较小. Multi-ported register file helps exploiting instruction-level and thread-level parallelism but brings area, energy and access time pressure. Oriented for superscalar and SMT processor, this paper give a method that is to manage a small auxiliary active value file (AVF) and selectively store physical register values in active cycle (during the time between production and last use). The AVF structure can share the register file access pressure and reduce the number of register ports, is simple to implement, and can filter some writes. It achieves significant energy savings with no impact on frequency and only small IPC loss.
出处 《计算机学报》 EI CSCD 北大核心 2008年第2期299-308,共10页 Chinese Journal of Computers
关键词 物理寄存器堆 寄存器重命名 寄存器生命周期 乱序执行 SMT physical register file register renaming register life time out-of-order execution SMT
  • 相关文献

参考文献1

二级参考文献17

  • 1Gowan M K, Biro L L, Jackson D B. Power considerations in the design of the Alpha 21264 microprocessor. In Proc.35th ACM/IEEE Design Automation Conf., San Francisco,California, June, 1998, pp.726-731.
  • 2Wilcox K, Manne S. Alpha processors: A history of power issues and a look to the future. In Cool-Chips Tutorial During 32nd Int. Syrup. Microarchitecture, Haifa, Israel, 1999.
  • 3Folegnani D, Gonzalez A. Energy-effective issue logic. In Proc.28th Int. Syrup. Computer Architecture, Goteberg, Sweden,June 30 July 4, 2001, pp.230-239.
  • 4Ernst D, Austin T. Efficient dynamic scheduling through tag elimination. In Proc. 29th Int. Syrup. Computer Architecture, Anchorage, Alaska, May 25-29, 2002, pp.37-46.
  • 5Albonesi D. Dynamic IPC/clock rage optimization. In Proc.25th Int. Syrup. Computer Architecture, Barcelona, Spain,June 27-July 1, 1998, pp.282 292.
  • 6Buyuktosunoglu A et al. A circuit level implementation of an adaptive issue queue for power-aware microprocessors. In Proc, 11th Great Lakes Syrup. VLSI Design, West Lafayette,Indiana, 2001, pp.73-78.
  • 7Dropsho S et al. Integrating adaptive on-chip storage structures for reduced dynamic power. In P.oc. 11th Int. Conf.Parallel Architectures and Compilation Techniques, Charlottesville, Virginia, Sept. 22-25, 2002, pp.141-152.
  • 8J Sharky, Dmitry V Ponomarev. Efficient instruction schedulers for SMT processors. In Proc. 12th Int. Syrup. High-Performance Computer Architecture, Austin, Texas, Feb. 11-15, 2006, pp.288 298.
  • 9Palacharla S, Jouppi N P, Smith J E. Complexity-effective superscalar processors. In Proc. 24th Int. Syrup. Computer Architecture, Denver, Colorado, June 2-4, 1997, pp.206-218.
  • 10Lebeck A R et al. A large, fast instruction window for tolerating cache misses. In Proc. 29th Int. Syrup. Computer Architecture, Anchorage, Alaska, May 25-29, 2002, pp.59-70.

共引文献2

同被引文献11

引证文献1

二级引证文献5

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部