11 July 2016 in IEEE Transactions on Nuclear Science
All Programmable System-on-Chip (APSoC) devices are designed to provide higher overall system performance and programmable flexibility at lower power consumption and costs. Although modern commercial APSoCs offer a plethora of advantages, they are prone to experience Single Event Upsets. We investigate the impact of using different system architectures on an APSoC in the overall system failure rate. We consider different memory organization, different communication schemes, and different computing modes. Results show that there are several choices of architectures and resources to be chosen to implement an application in an APSoC, but there are logic resources that can increase or decrease the vulnerability of the entire system to failures in the application execution context.
18 December 2017 in IEEE Transactions on Nuclear Science
In this paper we experimentally and analytically evaluate the reliability of two state-of-the-art neural networks for linear regression and pattern recognition (Multi-Layer Perceptron and Single-Layer Perceptron) implemented in a System on Chip composed of an FPGA and a microprocessor. We have considered, for each neural network, three different activation function complexities, to evaluate how the implementation affects FPGAs reliability. As we show in the paper, higher complexity increases the exposed area but reduces the probability of one failure to impact the network output. Additionally, we propose to distinguish between critical and tolerable errors in artificial neural networks. Experiments using a controlled heavy-ions beam show that, for both networks, only about 30% of the observed output errors actually affect the outputs correctness. We identify the causes of critical errors through fault injection, and found that faults in initial layers are more likely to significantly affect the output.
17 December 2015 in IEEE Transactions on Nuclear Science
Radiation effects such as soft errors are the major threat to the reliability of SRAM-based FPGAs. This work analyzes the effectiveness in correcting soft errors of a novel scrubbing technique using internal frame redundancy called Frame-level Redundancy Scrubbing (FLR-scrubbing). This correction technique can be implemented in a coarse grain TMR design. The FLR-scrubbing technique was implemented on a mid-size Xilinx Virtex-5 FPGA device used as a case study. The FLR-scrubbing technique was tested under neutron radiation and fault injection. Implementation results demonstrated minimum area and energy consumption overhead when compared to other techniques. The time to repair the fault is also improved by using the Internal Configuration Access Port (ICAP). Neutron radiation test results demonstrated that the proposed technique is suitable for correcting accumulated SEUs and MBUs.
17 July 2017 in IEEE Transactions on Nuclear Science
FKFernanda Lima Kastensmidt
Fernanda Lima Kastensmidt
+4
This paper shows the impact of low linear energy transfer heavy ions on the reliability of 28-nm Bulk static random access memory (RAM) cells from Artix-7 field-programmable gate array. Irradiation tests on the ground showed significant differences in the multiple bit upset cross section of configuration RAM and block RAM memory cells under various angles of incidence and rotation of the device. Experimental data are analyzed at transistor level by using the single-event effect prediction tool called multiscale single-event phenomenon prediction platform coupled with SPICE simulations.
12 July 2016 in IEEE Transactions on Nuclear Science
ARM processors are leaders in embedded systems, delivering high-performance computing, power efficiency, and reduced cost. For this reason, there is a relevant interest for its use in the aerospace industry. However, the use of sub-micron technologies has increased the sensitivity to radiation-induced transient faults. Thus, the mitigation of soft errors has become a major concern. Software-Implemented Hardware Fault Tolerance (SIHFT) techniques are a low-cost way to protect processors against soft errors. On the other hand, they cause high overheads in the execution time and memory, which consequently increase the energy consumption. In this work, we implement a set of software techniques based on different redundancy and checking rules. Furthermore, a low-overhead technique to protect the program execution flow is included. Tests are performed using the ARM Cortex-A9 processor. Simulated fault injection campaigns and radiation test with heavy ions have been performed. Results evaluate the trade-offs among fault detection, execution time, and memory footprint. They show significant improvements of the overheads when compared to previously reported techniques.