SyRA: Early System Reliability Analysis for Cross-Layer Soft Errors Resilience in Memory Arrays of Microprocessor Systems

Abstract

Cross-layer reliability is becoming the preferred solution when reliability is a concern in the design of a microprocessor-based system. Nevertheless, deciding how to distribute the error management across the different layers of the system is a very complex task that requires the support of dedicated frameworks for cross-layer reliability analysis. This paper proposes SyRA, a system-level cross-layer early reliability analysis framework for radiation induced soft errors in memory arrays of microprocessor-based systems. The framework exploits a multi-level hybrid Bayesian model to describe the target system and takes advantage of Bayesian inference to estimate different reliability metrics. SyRA implements several mechanisms and features to deal with the complexity of realistic models and implements a complete toolchain that scales efficiently with the complexity of the system. The simulation time is significantly lower than micro-architecture level or RTL fault-injection experiments with an accuracy high enough to take effective design decisions. To demonstrate the capability of SyRA, we analyzed the reliability of a set of microprocessor-based systems characterized by different microprocessor architectures (i.e., Intel x86, ARM Cortex-A15, ARM Cortex-A9) running both the Linux operating system or bare metal in the presence of single bit upsets caused by radiation induced soft errors. Each system under analysis executes different software workloads both from benchmark suites and from real applications.

Flodam: Cross-Layer Reliability Analysis Flow for Complex Hardware Designs

“…Although these two categories are complementary, few works consider their combination in order to accurately analyse the impact of the radiation from the environment on the hardware design and application workload under study. A recent cross-layer approach considers technology, circuit, hardware and application layers based on Bayesian models for single-bit faults in memory components [15]. However, multiple-bit faults and combinational logic faults cannot be neglected and should be included in the reliability analysis.…”

ReCo1: A Fault resilient technique of Correlation Sensitive Stochastic Designs

“…Using PTM, we showed that though the accuracy of an adder using MUX is controlled by select inputs, they must be taken in order with inputs to reduce error. Nevertheless to mention that majority of works are aligned towards reliability assessment and analysis [7,8].…”

On the Analysis of Real-time Operating System Reliability in Embedded Systems

“…Indeed, dependability can be reduced in many ways: without taking into consideration errors due to design, hardware or software bugs, issues occurred during fabrication, intentional tampering, or many other external events that can affect this property during the lifetime of the application. Due to the interaction of the circuit with the surrounding environment memory bit-flips, signal degradation, data loss, and permanent damage to the physical circuit may happen [2]. Usually, the system dependability can be achieved with a deep analysis of the system weaknesses, and then with the implementation of mitigation techniques that allow to reduce or completely remove them.…”