Abstract
In 3D NoC-based many-core systems, DRAM accesses behave differently due to their different communication distances and the latency gap of different DRAM accesses becomes bigger as the network size increases, which leads to unfair DRAM access performance among different nodes. This phenomenon may lead to high latencies for some DRAM accesses that become the performance bottleneck of the system. The paper addresses the DRAM access fairness problem in 3D NoC-based many-core systems by narrowing the latency difference of DRAM accesses as well as reducing the maximum latency. Firstly, the latency of a round-trip DRAM access is modeled and the factors causing DRAM access latency difference are discussed in detail. Secondly, the DRAM access fairness is further quantitatively analyzed through experiments. Thirdly, we propose to predict the network latency of round-trip DRAM accesses and use the predicted round-trip DRAM access time as the basis to prioritize the DRAM accesses in DRAM interfaces so that the DRAM accesses with potential high latencies can be transferred as early and fast as possible, thus achieving fair DRAM access. Experiments with synthetic and application workloads validate that our approach can achieve fair DRAM access and outperform the traditional First-Come-First-Serve (FCFS) scheduling policy and the scheduling policies proposed by reference [7] and [24] in terms of maximum latency, Latency Standard Deviation (LSD)1 and speedup. In the experiments, the maximum improvement of the maximum latency, LSD, and speedup are 12.8%, 6.57%, and 8.3% respectively. Besides, our proposal brings very small extra hardware overhead (<0.6%) in comparison to the three counterparts.
- Dennis Abts, Natalie D. Enright Jerger, John Kim, Dan Gibson, and Mikko H. Lipasti. 2009. Achieving Predictable Performance Through Better Memory Controller Placement in Many-core CMPs. In Proceedings of the 36th Annual International Symposium on Computer Architecture (ISCA’09). ACM, New York, NY, USA, 451--461. Google Scholar
Digital Library
- Shuming Chen, Yaohua Wang, Sheng Liu, Jianghua Wan, Haiyan Chen, Hengzhu Liu, Kai Zhang, Xiangyuan Liu, and Xi Ning. 2014. FT-Matrix: A Coordination-Aware Architecture for Signal Processing. IEEE Micro 34, 6 (Nov 2014), 64--73.Google Scholar
Cross Ref
- Synopsys Company. 2017. DesignWare DDR IP Solutions. (2017). https://www.synopsys.com/designware-ip/interface-ip/ddrn.html.Google Scholar
- Vinodh Cuppu, Bruce Jacob, Brian Davis, and Trevor Mudge. 1999. A Performance Comparison of Contemporary DRAM Architectures. In Proceedings of the 26th Annual International Symposium on Computer Architecture (ISCA’99). IEEE Computer Society, Washington, DC, USA, 222--233. Google Scholar
Digital Library
- William James Dally. 1992. Virtual-channel flow control. IEEE Transactions on Parallel and Distributed Systems 3, 2 (March 1992), 194--205. Google Scholar
Digital Library
- William James Dally and Brian Patrick Towles. 2004. Principles and Practices of Interconnection Networks. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA. Google Scholar
Digital Library
- Masoud Daneshtalab, Masoumeh Ebrahimi, Juha Plosila, and Hannu Tenhunen. 2013. CARS: Congestion-aware Request Scheduler for Network Interfaces in NoC-based Manycore Systems. In Proceedings of the Conference on Design, Automation and Test in Europe (DATE’13). EDA Consortium, San Jose, CA, USA, 1048--1051. Google Scholar
Digital Library
- Ronald Dreslinski, David Fick, Bharan Giridhar, Gyouho Kim, Sangwon Seo, Matthew Fojtik, Sudhir Satpathy, Yoonmyung Lee, Daeyeon Kim, Nurrachman Liu, Michael Wieckowski, Gregory Chen, Dennis Sylvester, David Blaauw, and Trevor Mudge. 2013. Centip3De: A 64-Core, 3D Stacked Near-Threshold System. IEEE Micro 33, 2 (March 2013), 8--16. Google Scholar
Digital Library
- Samsung Electronics. 2015. DDR I, II, and III. Device Operations and Timing Diagram. (October 2015). http://www.samsung.com/global/business/semiconductor.Google Scholar
- David Fick, Ronald G. Dreslinski, Bharan Giridhar, Gyouho Kim, Sangwon Seo, Matthew Fojtik, Sudhir Satpathy, Yoonmyung Lee, Daeyeon Kim, Nurrachman Liu, Michael Wieckowski, Gregory Chen, Trevor Mudge, David Blaauw, and Dennis Sylvester. 2013. Centip3De: A Cluster-Based NTC Architecture With 64 ARM Cortex-M3 Cores in 3D Stacked 130 nm CMOS. IEEE Journal of Solid-State Circuits 48, 1 (Jan 2013), 104--117.Google Scholar
Cross Ref
- David Fick, Ronald G. Dreslinski, Bharan Giridhar, Gyouho Kim, Sangwon Seo, Matthew Fojtik, Sudhir Satpathy, Yoonmyung Lee, Daeyeon Kim, Nurrachman Liu, Michael Wieckowski, Gregory Chen, Trevor Mudge, Dennis Sylvester, and David Blaauw. 2012. Centip3De: A 3930 DMIPS/W configurable near-threshold 3D stacked system with 64 ARM Cortex-M3 cores. In 2012 IEEE International Solid-State Circuits Conference. IEEE Press, Washington, DC, USA, 190--192.Google Scholar
Cross Ref
- ITU. 2003. ITU-T Recommendation and Final Draft International Standard of Joint Video Specification (ITU-T H.264, ISO/IEC 14496-10 AVC). ITU Press, USA.Google Scholar
- Philip Jacob, Aamir Zia, Okan Erdogan, Paul M. Belemjian, Jin-Woo Kim, Michael Chu, and Russell P. Kraft. 2009. Mitigating Memory Wall Effects in High-Clock-Rate and Multicore CMOS 3-D Processor Memory Stacks. Proc. IEEE 97, 1 (Jan 2009), 108--122.Google Scholar
Cross Ref
- Wooyoung Jang and David Z. Pan. 2010. An SDRAM-aware Router for Networks-on-chip. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 29, 10 (Oct. 2010), 1572--1585. Google Scholar
Digital Library
- Wooyoung Jang and David Z. Pan. 2011. Application-Aware NoC Design for Efficient SDRAM Access. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 30, 10 (Oct. 2011), 1521--1533. Google Scholar
Digital Library
- Daehyun Kim, Krit Athikulwongse, Michael B. Healy, Mohammad M. Hossain, Moongon Jung, and Ilya Khorosh. 2012. 3D-MAPS: 3D Massively parallel processor with stacked memory. In 2012 IEEE International Solid-State Circuits Conference. IEEE Press, Washington, DC, USA, 188--190.Google Scholar
Cross Ref
- Dongki Kim, Sungjoo Yoo, and Sunggu Lee. 2010. A Network Congestion-Aware Memory Controller. In Proceedings of the 2010 Fourth ACM/IEEE International Symposium on Networks-on-Chip (NOCS’10). IEEE Computer Society, Washington, DC, USA, 257--264. Google Scholar
Digital Library
- Amit Kumar, Li-Shiuan Peh, Partha Kundu, and Niraj K. Jha. 2007. Express Virtual Channels: Towards the Ideal Interconnection Fabric. In Proceedings of the 34th Annual International Symposium on Computer Architecture (ISCA’07). ACM, New York, NY, USA, 150--161. Google Scholar
Digital Library
- Onur Mutlu and Thomas Moscibroda. 2007. Stall-Time Fair Memory Access Scheduling for Chip Multiprocessors. In Proceedings of the 40th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO 40). IEEE Computer Society, Washington, DC, USA, 146--160. Google Scholar
Digital Library
- Vasilis F. Pavlidis and Eby G. Friedman. 2007. 3D Topologies for Networks-on-chip. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 15, 10 (Oct. 2007), 1081--1090. Google Scholar
Digital Library
- Li-Shiuan Peh and Natalie Enright Jerger. 2009. On-Chip Networks (1st ed.). Morgan and Claypool Publishers, Washington, DC, USA. Google Scholar
Digital Library
- Tejasi Pimpalkhute and Sudeep Pasricha. 2014. NoC Scheduling for Improved Application-Aware and Memory-Aware Transfers in Multi-core Systems. In Proceedings of the 2014 27th International Conference on VLSI Design and 2014 13th International Conference on Embedded Systems (VLSID’14). IEEE Computer Society, Washington, DC, USA, 234--239. Google Scholar
Digital Library
- Scott Rixner, William J. Dally, Ujval J. Kapasi, Peter Mattson, and John D. Owens. 2000. Memory Access Scheduling. In Proceedings of the 27th Annual International Symposium on Computer Architecture (ISCA’00). ACM, New York, NY, USA, 128--138. Google Scholar
Digital Library
- Akbar Sharifi, Emre Kultursay, Mahmut Kandemir, and Chita R. Das. 2012. Addressing End-to-End Memory Access Latency in NoC-Based Multicores. In Proceedings of the 2012 45th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO-45). IEEE Computer Society, Washington, DC, USA, 294--304. Google Scholar
Digital Library
- Matthew R. Wordeman, Joel Silberman, Gary W. Maier, and Michael Scheuermann. 2012. A 3D system prototype of an eDRAM cache stacked over processor-like logic using through-silicon vias. In 2012 IEEE International Solid-State Circuits Conference. IEEE Press, Washington, DC, USA, 186--187.Google Scholar
Cross Ref
- George L. Yuan, Ali Bakhoda, and Tor M. Aamodt. 2009. Complexity Effective Memory Access Scheduling for Many-core Accelerator Architectures. In Proceedings of the 42Nd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO 42). ACM, New York, NY, USA, 34--44. Google Scholar
Digital Library
- Guangfei Zhang, Huandong Wang, Xinke Chen, and Peng Li. 2012. Fair Memory Access Scheduling for Quality of Service Guarantees via Service Curves. In Proceedings of the 2012 IEEE 10th International Symposium on Parallel and Distributed Processing with Applications (ISPA’12). IEEE Computer Society, Washington, DC, USA, 174--181. Google Scholar
Digital Library
- Yuang Zhang, Li Li, Zhonghai Lu, and Minglun Gao. 2014. Performance and network power evaluation of tightly mixed SRAM NUCA for 3D Multi-core Network on Chips. In 2014 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE Press, Washington, DC, USA, 1961--1964.Google Scholar
Cross Ref
- Yuang Zhang, Li Li, Zhonghai Lu, Axel Jantsch, Minglun Gao, Hongbing Pan, and Feng Han. 2014. A Survey of Memory Architecture for 3D Chip Multi-processors. Microprocess. Microsyst. 38, 5 (July 2014), 415--430. Google Scholar
Digital Library
Index Terms
Round-trip DRAM Access Fairness in 3D NoC-based Many-core Systems
Recommendations
Distributed fair DRAM scheduling in network-on-chips architecture
Memory access scheduling is an effective manner to improve performance of Chip Multi-Processors (CMPs) by taking advantage of the timing characteristics of a DRAM. A memory access scheduler can subdivide resources utilization (banks and rows) to ...
Memory access pattern-aware DRAM performance model for multi-core systems
ISPASS '11: Proceedings of the IEEE International Symposium on Performance Analysis of Systems and SoftwareThe DRAM latency modeling is complex because most chips contain row-buffers and multiple banks to exploit patterns of DRAM accesses. As a result, the latency of DRAM access not only depends on the circuit timing parameters but also memory access ...
Agent-Based Memory Access for Many-Core CMPs
ISPDC '14: Proceedings of the 2014 IEEE 13th International Symposium on Parallel and Distributed ComputingThe trend of increasing on-chip core counts and integrating memory controllers (MCs) makes core-to-memory communication a major obstacle in scaling memory access performance for many-core CMPs. Unmanaged on-chip traffic for long-distance memory accesses,...






Comments