The present paper examines reliability design of computing systems. Systems with strict reliability requirements utilize structural, informational and timing redundancy, as well as recovery techniques and facilities for life data collection and analysis. In this paper we present a reliability design strategy, based on analysis of existent periodic information recovery and implementation of additional forced periodicity in computing processes at various levels of design hierarchy. We present new analytical models for reliability evaluation and a new approach to synthesis, based on operation periodicity in units of the system. The key term in proposed models is a unit’s information recovery period. We show that there is a strong dependency between reliability of the system and lengths of these periods. Models are applicable both for triplicated units and for units with error-correcting codes and n-modular redundancy
Keywords: fault tolerance, hardware redundancy, reliability design, soft error, self-recovery.
1. Polian, I., Hayes, J.P. Selective hardening: toward cost-effective error tolerance.
IEEE Design and test of computers, 2011, 54–62.
2. Platzker, D., Mohsen, E. Using FPGA synthesis to protect against radiation effects
and soft errors. Mentor graphics white paper, 2011.
3. Maximenko, S.L., Melekhin, V.F. Reliability Analysis of Digital Devices with
Structural Redundancy and Periodic Operational State Recovery of Functional Nodes. Information and Control Systems, 2013, 3, 16–22.
4. Maximenko, S.L., Melekhin, V.F. Analysis of Reliability of Functional Nodes of Digital VLSI Circuits with Structural Redundancy and Periodic Operational State Recovery. Information and Control Systems, 2013, 2, 18–23.
5. Edmonds, L.D. et al. An introduction to space radiation effects on microelectronics.
JPL publication 00-06, 2000.
6. Rao, R.R., Chopra, K., Blaauw, D.T., Sylvester, D.M. Computing the soft error rate of a combinational logic circuit using parameterized descriptors. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2006, vol. 26 (3), 468–479.
7. Hass, K.J., Ambles, J.W. Single event transients in deep submicron CMOS. 42nd Midwest Symposium on Circuits and Systems, 2000, vol. 1, 122–125.
8. Benedetto, J.M. et al. Variation of digital SET pulse width and the implications for single event hardening of advanced CMOS processes. IEEE transactions on Nuclear Science, 2005, vol. 52, 2114–2119.
9. Glukhih, M.I. Development of synthesis methods for application specific information and control systems with structural redundancy. PhD Thesis. 2006.
10. Abraham, J.A., Siewiorek, D.P. An algorithm for the accurate reliability evaluation of triple modular redundancy networks. IEEE Transactions on Computers, 1974, vol. C-23 (7), 682–692.
