DocumentCode :
2950070
Title :
Optimizing communication and capacity in a 3D stacked reconfigurable cache hierarchy
Author :
Madan, Niti ; Zhao, Li ; Muralimanohar, Naveen ; Udipi, Aniruddha ; Balasubramonian, Rajeev ; Iyer, Ravishankar ; Makineni, Srihari ; Newell, Donald
Author_Institution :
Sch. of Comput., Univ. of Utah, Salt Lake City, UT
fYear :
2009
fDate :
14-18 Feb. 2009
Firstpage :
262
Lastpage :
274
Abstract :
Cache hierarchies in future many-core processors are expected to grow in size and contribute a large fraction of overall processor power and performance. In this paper, we postulate a 3D chip design that stacks SRAM and DRAM upon processing cores and employs OS-based page coloring to minimize horizontal communication of cache data. We then propose a heterogeneous reconfigurable cache design that takes advantage of the high density of DRAM and the superior power/delay characteristics of SRAM to efficiently meet the working set demands of each individual core. Finally, we analyze the communication patterns for such a processor and show that a tree topology is an ideal fit that significantly reduces the power and latency requirements of the on-chip network. The above proposals are synergistic: each proposal is made more compelling because of its combination with the other innovations described in this paper. The proposed reconfigurable cache model improves performance by up to 19% along with 48% savings in network power.
Keywords :
DRAM chips; SRAM chips; cache storage; integrated circuit design; multiprocessing systems; network topology; network-on-chip; reconfigurable architectures; trees (mathematics); 3D stacked reconfigurable cache hierarchy; DRAM chip; OS-based page coloring; SRAM chip; chip design; horizontal communication; multicore processor; network-on-chip; tree topology; Bandwidth; Chip scale packaging; Delay; Network topology; Network-on-a-chip; Proposals; Random access memory; Space technology; Stacking; Technological innovation; SRAM/DRAM cache reconfiguration; cache and memory hierarchy; multi-core processors; non-uniform cache architecture (NUCA); on-chip networks; page coloring;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
High Performance Computer Architecture, 2009. HPCA 2009. IEEE 15th International Symposium on
Conference_Location :
Raleigh, NC
ISSN :
1530-0897
Print_ISBN :
978-1-4244-2932-5
Type :
conf
DOI :
10.1109/HPCA.2009.4798261
Filename :
4798261
Link To Document :
بازگشت