Skip to main content
Log in

GPUs Neutron Sensitivity Dependence on Data Type

  • Published:
Journal of Electronic Testing Aims and scope Submit manuscript

Abstract

Graphics Processing Units are very prone to be corrupted by neutrons. Experimental results obtained irradiating the GPU with high energy neutrons show that the input data type has a strong influence on the neutron-induced error-rate of the executed algorithms. Moreover, when operations are performed using floating-point data, the probabilities to be corrupted are very different for the mantissa, the exponent or the sign. We investigate the occurrences of errors in the different positions, evaluating the related effects on the result precision. The reported results and the architecture analysis demonstrate that under radiation, whenever possible, one should favor floating-point arithmetic, which is both more reliable and potentially easier to protect than the integer one.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

References

  1. Baumann R C (2005) Radiation-induced soft errors in advanced semiconductor technologies. IEEE Trans Dev Mater Reliab 5(3):305–316

    Article  MathSciNet  Google Scholar 

  2. Di Carlo S et al (2013) A software-based self test of CUDA fermi GPUs. IEEE Eur Test Symp: 3338

  3. Haque I S, Pande V S (2010) Hard data on soft errors: a large- scale assessment of real-world error-rates in GPGPU. In: Proceedings of the IEEE/ACM international conference on cluster, cloud and grid computing, pp 691–696

  4. Kim S-H et al (2007) A low power and highly reliable 400Mbps mobile DDR SDRAM with on-chip distributed ECC. In: Proceedings of IEEE Asian solid-state circuits conference. pp 34–37

  5. Kirk DB, Hwo WW Programming massively parallel processors. MK Publishers

  6. Kruger J, Westermann R (2003) Linear algebra operators for GPU implementation of numerical algorithms. ACM Trans Graph 3(22):908–916

    Article  Google Scholar 

  7. Liepe J, Barnes C, Cule E, Erguler K, Kirk P, Toni T, Stumpf M P H (2012) ABC-SysBio-approximate Bayesian computation in Python with GPU support. Bioinformatics 26(14):1797–1799

    Article  Google Scholar 

  8. Lindholm E, Nickolls J, Oberman S, Montrym J (2008) NVIDIA Tesla: a unified graphics and computing architecture. IEEE MICRO 28(2):39–55

    Article  Google Scholar 

  9. Nguyen HT, Yagil Y, Seifert N, Reitsma M (2005) Chip-level soft error estimation method. IEEE Trans Dev Mater Reliab 5(3):356–381

    Article  Google Scholar 

  10. NVIDIA GeForce GTX 480/470/465 GPU Datasheet

  11. Owens JD, Houston M, Luebke D, Green S, Stone JE, Phillips JC (2008) GPU Computing. Proc IEEE 96(5):879–899

    Article  Google Scholar 

  12. Pilla LL, Rech P, Silvestri F, Frost C, Navaux P O A, Sonza Reorda M, Carro L (2014) Software-based hardening strategies for neutron sensitive FFT algorithms on GPUs. IEEE Trans Nucl Sci 99:1–7

    Article  Google Scholar 

  13. Rech P, Aguiar C, Frost C, Carro L (2012) Neutron radiation test of graphic processing units. In: Proceedings IEEE IOLTS

  14. Rech P, Aguiar C, Ferreira R, Silvestri M, Griffoni A, Frost C, Carro L (2012) Neutron-induced soft error in graphic processing units. In: Proceedings of IEEE REDW

  15. Rech P, Aguiar C, Frost C, Carro L An efficient and experimentally tuned software-based hardening strategy for matrix multiplication on GPUs. IEEE Trans Nucl Sci 60(4):2797–2804

  16. Sheaffer JW, Luebke DP, Skadron K (2007) A hardware redundancy and recovery mechanism for reliable scientific computation on graphics processors. In: Proceedings of the ACM SIGGRAPH symposium on graphics hardware (GH). pp 55–64

  17. Seifert N, Xiaowei Z, Massengill L W Impact of scaling on soft-error-rates in commercial microprocessors, vol 46

  18. Shi G, Enos J, Showerman M, Kindratenko V (2009) On testing GPU memory for hard and soft errors. In: Proceedings of the symposium on application accelerators in high-performance computing (SAAHPC)

  19. Violante M et al A New hardware/software platform and a new 1/E neutron source for soft error studies: testing FPGAs at the ISIS facility. IEEE TNS Nucl Sci 54(4):1184–1189

  20. Wang NJ, Quek J, Rafacz TM, Patel SJ (2004) Characterizing the effects of transient faults on a high-performance processor pipeline. In: Proceedings of the IEEE international conference on dependable systems and networks (DSN). pp 61–70

Download references

Acknowledgments

This work is supported by CAPES foundation of the Ministry of Education, CNPq research council of the Ministry of Science and Technology, and FAPERGS research agency of the State of Rio Grande do Sul, Brazil. Experiments were performed in ISIS, Rutherford Appleton Laboratories, Didcot, UK and founded by the Science and Technology Faculty Council (STFC), UK.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to P. Rech.

Additional information

Responsible Editor: F. L. Vargas

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Rech, P., Frost, C. & Carro, L. GPUs Neutron Sensitivity Dependence on Data Type. J Electron Test 30, 307–316 (2014). https://doi.org/10.1007/s10836-014-5456-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10836-014-5456-6

Keywords

Navigation