Write-variation aware alternatives to replace SRAM buffers with non-volatile buffers in on-chip interconnects

被引:2
作者
Rani, Khushboo [1 ]
Kapoor, Hemangee K. [1 ]
机构
[1] IIT Guwahati, Dept Comp Sci & Engn, Gauhati 781039, Assam, India
关键词
integrated circuit design; network-on-chip; buffer circuits; CMOS memory circuits; SRAM chips; cache storage; MRAM devices; SRAM buffers; nonvolatile buffers; on-chip interconnects; CMOS technology; multiple processors; NoC interconnects; leakage power consumption; NoC buffers; nonvolatile spin transfer torque random access memory-based buffers; STT-RAM technology; write-variations; virtual channel allocation policies; VC allocation policies; iso-area-based alternatives; pure STT-RAM buffers; SRAM technology; write variation; HYBRID CACHE; POWER; AREA; PERFORMANCE; EFFICIENT; ROUTERS; ENERGY; MEMORY; MODEL;
D O I
10.1049/iet-cdt.2019.0039
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
With the advancement in CMOS technology and multiple processors on the chip, communication across these cores is managed by a network-on-chip (NoC). Power and performance of these NoC interconnects have become a significant factor.The authors aim to reduce the leakage power consumption of NoC buffers by the use of non-volatile spin transfer torque random access memory (STT-RAM)-based buffers. STT-RAM technology has the advantages of high density and low leakage but suffers from low endurance. This low endurance has an impact on the lifetime of the router on the whole due to unwanted write-variations governed by virtual channel (VC) allocation policies. Here various VC allocation policies that help the uniform distribution of the writes across the buffers are proposed. Iso-capacity and iso-area-based alternatives to replace SRAM buffers with STT-RAM buffers are also presented. Pure STT-RAM buffers, however, impact the network latency. To mitigate this, a hybrid variant of the proposed policies which uses alternative VCs made of SRAM technology in the case of heavy network traffic is proposed. Experimental evaluation of full system simulation shows that proposed policies reduce the write variation by 99% and improve lifetime by 3.2 times and 1093 times, respectively. Also a 55.5% gain in the energy delay product is obtained.
引用
收藏
页码:481 / 492
页数:12
相关论文
共 40 条
[1]   GARNET: A Detailed On-Chip Network Model inside a Full-System Simulator [J].
Agarwal, Niket ;
Krishna, Tushar ;
Peh, Li-Shiuan ;
Jha, Niraj K. .
ISPASS 2009: IEEE INTERNATIONAL SYMPOSIUM ON PERFORMANCE ANALYSIS OF SYSTEMS AND SOFTWARE, 2009, :33-42
[2]   Spin-Transfer Torque Magnetic Random Access Memory (STT-MRAM) [J].
Apalkov, Dmytro ;
Khvalkovskiy, Alexey ;
Watts, Steven ;
Nikitin, Vladimir ;
Tang, Xueti ;
Lottis, Daniel ;
Moon, Kiseok ;
Luo, Xiao ;
Chen, Eugene ;
Ong, Adrian ;
Driskill-Smith, Alexander ;
Krounbi, Mohamad .
ACM JOURNAL ON EMERGING TECHNOLOGIES IN COMPUTING SYSTEMS, 2013, 9 (02)
[3]  
Arcaro S, 2014, INT DES TEST SYMP, P67, DOI 10.1109/IDT.2014.7038589
[4]  
Bell S., 2008, P 2008 IEEE INT SOL, P88, DOI DOI 10.1109/ISSCC.2008.4523070
[5]  
Binkert Nathan, 2011, Computer Architecture News, V39, P1, DOI 10.1145/2024716.2024718
[6]  
Bohr M., 2007, IEEE Solid-State Circuits Soc. Newsl, V12, P11, DOI [DOI 10.1109/N-SSC.2007.4785534, DOI 10.1109/NSSC.2007.4785534, 10.1109/N-SSC.2007.4785534]
[7]   POSTER: Fly-Over: A Light-Weight Distributed Power-Gating Mechanism for Energy-Efficient Networks-on-Chip [J].
Boyapati, Rahul ;
Huang, Jiayi ;
Wang, Ningyuan ;
Kim, Kyung Hoon ;
Yum, Ki Hwan ;
Kim, Eun Jung .
2016 INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURE AND COMPILATION TECHNIQUES (PACT), 2016, :413-414
[8]  
Chen LZ, 2015, INT S HIGH PERF COMP, P378, DOI 10.1109/HPCA.2015.7056048
[9]  
Chen YT, 2012, DES AUT TEST EUROPE, P45
[10]  
Ding L, 2012, PROCEEDINGS OF ISCRAM ASIA 2012 CONFERENCE ON INFORMATION SYSTEMS FOR CRISIS RESPONSE AND MANAGEMENT, P201