dsReliM: power-constrained reliability management in dark-silicon many-core chips under process variations
Authors: 
Mohammad Salehi, Muhammad Shafique, Florian Kriebel, Semeen Rehman, Mohammad Khavari Tavana, Alireza Ejlali, Jörg HenkelAuthors Info & Claims
Mohammad Salehi, Muhammad Shafique, Florian Kriebel, Semeen Rehman, Mohammad Khavari Tavana, Alireza Ejlali, Jörg HenkelAuthors Info & ClaimsPages 75 - 82
Abstract
Due to the tight power envelope, in the future technology nodes it is envisaged that not all cores in a many-core chip can be simultaneously powered-on (at full performance level). The power-gated cores are referred to as Dark Silicon. At the same time, growing reliability issues due to process variations and soft errors challenge the cost-effective deployment of future technology nodes. This paper presents a reliability management system for Dark Silicon chips (dsReliM) that optimizes for reliability of on-chip systems while jointly accounting for soft errors, process variations and the thermal design power (TDP) constraint. Towards the TDP-constrained reliability optimization, dsReliM leverages multiple reliable application versions that can potentially execute on different cores with frequency variations and supporting differenst voltage-frequency levels, thus facilitating distinct power, reliability and performance tradeoffs at run time. Experiments show that our dsReliM system provides up to 20% reliability improvements under different TDP constraints when compared to a state-of-the-art technique. Also, compared to an ideal-case optimal solution, dsReliM deviates up to 2.5% in terms of reliability efficiency, but speeds up the reliability management decision time by a factor of up to 3100.
References
[1]
International technology roadmap for semiconductors, http://public.itrs.net/reports.html.
[2]
H. Esmaeilzadeh, E. Blem, R. St. Amant, K. Sankaralingam, and D. Burger. Dark silicon and the end of multicore scaling. In International Symposium on Computer Architecture (ISCA), pp. 365--376, 2011.
[3]
M. B. Taylor. Is dark silicon useful?: harnessing the four horsemen of the coming dark silicon apocalypse. In ACM/EDAC/IEEE Design Automation Conference (DAC), pp. 1131--1136, 2012.
[4]
M. Shafique, S. Garg, J. Henkel, and D. Marculescu. The eda challenges in the dark silicon era: Temperature, reliability, and variability perspectives. In ACM/EDAC/IEEE Design Automation Conference (DAC), pp. 1--6, 2014.
[5]
M. Shafique, S. Garg, T. Mitra, S. Parameswaran, and J. Henkel. Dark silicon as a challenge for hardware/software co-design: invited special session paper. In IEEE CODES+ISSS 2014.
[6]
R. Teodorescu and J. Torrellas. Variation-Aware Application Scheduling and Power Management for Chip Multiprocessors. In International Symposium on Computer Architecture (ISCA), pp. 363--374, 2008.
[7]
M. Shafique, D. Gnad, S. Garg, J. Henkel. Variability-Aware Dark Silicon Management in On-Chip Many-Core Systems. In IEEE/ACM Design, Automation and Test in Europe Conference (DATE'15), France, 9--13 March, 2015.
[8]
N. Kapadia, S. Pasricha. VARSHA: Variation and Reliability-Aware Application Scheduling with Adaptive Parallelism in the Dark-Silicon Era In IEEE/ACM Design, Automation and Test in Europe Conference (DATE'15), France, 9--13 March, 2015.
[9]
R. Zheng, J. Velamala, V. Reddy, V. Balakrishnan, E. Mintarno, S. Mitra, S. Krishnan, Y. Cao. Circuit aging prediction for low-power operation. In Custom Integrated Circuits Conference (CICC), pp.427--430, 2009.
[10]
J. Henkel, L. Bauer, N. Dutt, P. Gupta, S. Nassif, M. Shafique, M. Tahoori, N. When. Reliable On-Chip Systems in the Nano-Era: Lessons Learnt and Future Trends. In ACM/EDAC/IEEE Design Automation Conference (DAC), pp. 1--10, 2013.
[11]
D. Brooks, R. P. Dick, R. Joseph, L. Shang. Power, Thermal, and Reliability Modeling in Nanometer-Scale Microprocessors, In IEEE Micro, pp.49--62, 2007.
[12]
T. D. Loveless, S. Jagannathan, T. Reece, J. Chetia, B. L. Bhuva, M. W. McCurdy, L. W. Massengill, S.-J. Wen, R. Wong, D. Rennie. Neutron- and proton-induced single event upsets for D- and DICE-flip/flop designs at a 40 nm technology node, IEEE Trans. Nucl. Sci., 58(3), pp. 1008--1014, June 2011.
[13]
S. S. Mukherjee, M. Kontz, S. K. Reinhardt. Detailed design and evaluation of redundant multithreading alternatives, In IEEE Int'l Symp. Comput. Arch. (ISCA), pp. 99--110, 2002.
[14]
A. Shye, T. Moseley, V. Janapa Reddi, J. Blomstedt, D. A. Connors. Using Process-Level Redundancy to Exploit Multiple Cores for Transient Fault Tolerance, In IEEE/IFIP Int'l Conf. Depend. Syst. Net. (DSN), pp. 297--306, 2007.
[15]
M. Salehi, M. K. Tavana, S. Rehman, F. Kriebel, M. Shafique, A. Ejlali, J. Henkel. DRVS: Power-Efficient Reliability Management through Dynamic Redundancy and Voltage Scaling under Variations. In ACM/IEEE Int'l Symp. Low Power Electron. Design (ISLPED), 2015.
[16]
K. K. Rangan, M. Powell, G.-Y. Wei, D. Brooks. Achieving Uniform Performance and Maximizing Throughput in the Presence of Heterogeneity, In IEEE Int'l Symp. High Perform. Comput. Arch. (HPCA), pp. 3--14, 2011.
[17]
S. Rehman, M. Shafique, F. Kriebel, J. Henkel. Reliable software for unreliable hardware: Embedded code generation aiming at reliability, In IEEE CODES+ISSS, pp. 237--246, 2011.
[18]
S. Rehman. A. Toma, F. Kriebel, M. Shafique, J.-J. Chen, J. Henkel. Reliable Code Generation and Execution on Unreliable Hardware under Joint Functional and Timing Reliability Considerations, In IEEE Real-Time Embed. Tech. App. Symp. (RTAS), pp. 273--282, 2013.
[19]
A. Ejlali, B. M. Al-Hashimi, and P. Eles. Low-Energy Stand-by-Sparing for Hard Real-Time Systems. IEEE Trans. Comput.-Aid. Des. Integr. Circuits Syst., 31 (3), 329--342.
[20]
D. Zhu, R. Melhem, D. Mossé. The effects of energy management on reliability in real-time embedded systems, In IEEE/ACM Int'l Conf. Comput. Aided Design (ICCAD), pp.35--40, 2004.
[21]
I. Koren, C. M. Krishna. Fault-Tolerant Systems. Morgan Kaufmann, Elsevier, San Francisco, CA, 2007.
[22]
F. Kriebel, S. Rehman, D. Sun, M. Shafique, and J. Henkel. ASER: Adaptive Soft Error Resilience for Reliability-Heterogeneous Processors in the Dark Silicon Era. In ACM/EDAC/IEEE Design Automation Conference (DAC), pp.1--6, 2014.
[23]
T. D. Burd, T. A. Pering, A. J. Stratakos, and R. Brodersen. A dynamic voltage scaled microprocessor system. IEEE J. Solid-State Circuits (JSSC), 35 (11), 1571--1580, 2000.
[24]
M. Salehi and A. Ejlali. A Hardware Platform for Evaluating Low-Energy Multiprocessor Embedded Systems Based on COTS Devices, IEEE Trans. Indust. Electron., 62(2), 2015.
[25]
"Gurobi optimizer." {Online}. Available: www.gurobi.com.
[26]
R. Karp. Reducibility among combinatorial problems. Springer, 1972.
[27]
J. Mitchell. Branch-and-cut algorithms for combinatorial optimization problems, in Handbook of Applied Optimization, 2002.
[28]
G. Liu, J. Park, and D. Marculescu. Dynamic Thread Mapping for High-Performance, Power-Efficient Heterogeneous Many-core Systems. In IEEE Int'l Conf. Comput. Design (ICCD), pp. 54--61, 6--9 Oct. 2013.
[29]
B. Raghunathan, Y. Turakhia, S. Garg, and D. Marculescu. Cherry-Picking: Exploiting Process Variations in Dark-Silicon Homogeneous Chip Multi-Processors, In IEEE/ACM Design, Automation and Test in Europe Conference (DATE'13), pp. 39--44, 2013.
[30]
P. Friedberg, Y. Cao, J. Cain, R. Wang, J. Rabaey, and C. Spanos. Modeling within-die spatial correlation effects for process-design co-optimization, In IEEE Int'l Symp. Quality Electron. Design (ISQED), pp. 516--521, 2005.
[31]
S. Herbert and D. Marculescu. Variation-Aware Dynamic Voltage/Frequency Scaling, In IEEE Int'l Symp. High Perform. Comput. Arch. (HPCA), pp. 301--312, 2009.
[32]
A. Srivastava, R. Bai, D. Blaauw, D. Sylvester. Modeling and analysis of leakage power considering within-die process variations, In ACM/IEEE Int'l Symp. Low Power Electron. Design (ISLPED), pp. 64--67, 2002.
[33]
Predictive Technology Model (PTM). available at: http://ptm.asu.edu/latest.html.
[34]
M. K. Tavana, A. Kulkarni, A. Rahimi, T. Mohsenin, and H. Homayoun. Energy-Efficient Mapping of Biomedical Applications on Domain-Specific Accelerator under Process Variation, In ACM/IEEE Int'l Symp. Low Power Electron. Design (ISLPED), 2014.
[35]
D. Rhodes and R. Dick. TGFF: Task Graphs for Free, In CODES/CASHE, 1998.
Index Terms
- dsReliM: power-constrained reliability management in dark-silicon many-core chips under process variations
Recommendations
Leveraging dark silicon to optimize networks-on-chip topology
This paper presents a reconfigurable network-on-chip (NoC) for many-core chip multiprocessors (CMPs) in the dark silicon era, where a considerable part of high-end chips cannot be powered up due to the power and bandwidth walls. Core specialization, ...
Shift sprinting: fine-grained temperature-aware NoC-based MCSoC architecture in dark silicon age
DAC '16: Proceedings of the 53rd Annual Design Automation ConferenceReliability is a critical feature of chip integration and unreliability can lead to performance, cost, and time-to-market penalties. Moreover, upcoming Many-Core System-on-Chips (MCSoCs), notably future generations of mobile devices, will suffer from ...
Efficient Mapping of Applications for Future Chip-Multiprocessors in Dark Silicon Era
The failure of Dennard scaling has led to the utilization wall that is the source of dark silicon and limits the percentage of a chip that can actively switch within a given power budget. To address this issue, a structure is needed to guarantee the ...
Comments
Information & Contributors
Information
Published In
Sponsors
- SIGBED: ACM Special Interest Group on Embedded Systems
- SIGDA: ACM Special Interest Group on Design Automation
- IEEE: Institute of Electrical and Electronics Engineers
- SIGMICRO: ACM Special Interest Group on Microarchitectural Research and Processing
Publisher
IEEE Press
Publication History
Published: 04 October 2015
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
Acceptance Rates
Overall Acceptance Rate 280 of 864 submissions, 32%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 106Total Downloads
- Downloads (Last 12 months)2
- Downloads (Last 6 weeks)1
Reflects downloads up to 17 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in