Classification of Resilience Techniques Against Functional Errors at Higher Abstraction Layers of Digital Systems

Psychou, Georgia; Rodopoulos, Dimitrios; Sabry, Mohamed M.; Gemmeke, Tobias; Atienza, David; Noll, Tobias G.; Catthoor, Francky

doi:10.1145/3092699

Cited by 5 publications

(1 citation statement)

References 126 publications

(85 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to the aging adaptation and mitigation techniques explained above, it is worth to note that there are other mitigation methodologies for digital systems at higher abstraction level for functional errors. Examples are hardware platform, and mapping and software platform techniques [175]. All the presented mitigation techniques focus mostly on logic circuits or systems (i.e., processor or architecture).…”

mentioning

confidence: 99%

Reliability Modeling and Mitigation for Embedded Memories

Agbo

Taouil

Hamdioui

2019

2019 IEEE International Test Conference (ITC)

View full text Add to dashboard Cite

Complementary Metallic Oxide Semiconductor (CMOS) technology scaling enhances the performance, transistor density, functionality, and reduces cost and power consumption. However, scaling causes significant reliability challenges both from a manufacturing and operational point of view. Obtaining reliable memories require accurate understanding of the impact of aging (such as Bias temperature instability (BTI)) on individual memory components and how they interact with each other. In this dissertation, two types of challenges are addressed, which are related to BTI aging and partially to mitigation schemes: one related to the aging of sense amplifier and another one to the aging of read path and write path. Analysis of aging impact on different memory sense amplifiers-The analysis of BTI impact on various memory sense amplifier (SA) designs was performed, while taking into account two BTI models (i.e., Atomistic and RD model), different technology nodes (i.e., 90, 65, 45, 32, 22, and 16 nm), and different workloads. First, the analysis and comparison of RD and Atomistic models impact on the SA were performed. The results show that the atomistic trap-based BTI model is more accurate than the RD model. Second, the investigation of BTI impact on the drain-input latch type SA for various technology nodes and supply voltages was performed. The result shows that as technology scales down, the impact of BTI on sensing delay increases, while the sensing voltage decreases, causing less robust and reliable memory sense amplifier. The result also shows that increase in supply voltage compensates the BTI degradation. Third, an accurate technique was proposed and characterized for the integral impact of BTI and voltage temperature variation on the memory standard latch type SA for various technology nodes and workloads. The results show that the degradation is strongly dependent on workload and temperature. Fourth, in addition to the latter, the impact of process variation at timezero was incorporated and analyzed. The results show that the SA sensing delay degradation is more significant at lower nodes and could lead to read failures at lower power supply. This reveals that there must be a tradeoff between performance and reliability. Fifth, an accurate methodology was proposed to quantify the impact of variability on the memory SA offset-voltage for both time-zero and time-dependent variability. The results show that the impact on the offset voltage specification is significant for aging time-dependent variability. Sixth, on top of the latter, the sensitivity of the SA and its failure rate were analyzed for five process corners (i.e., Nominal, Fast-Fast, Fast-Slow, Slow-Fast, and Slow-Slow). The results show that balanced workloads result in a significant low offset voltage specification. Finally, the impact of aging was analyzed and compared, while considering different supply voltages, temperatures, and SA designs. The results show that the High Performance SA degrades faster than other SA types, irrespective of the workloa...

show abstract

mentioning

confidence: 99%

Reliability Modeling and Mitigation for Embedded Memories

Agbo

Taouil

Hamdioui

2019

2019 IEEE International Test Conference (ITC)

View full text Add to dashboard Cite

show abstract

Preliminaries

Koren¹,

Krishna²

2021

Fault-Tolerant Systems

View full text Add to dashboard Cite

A Closed-Loop Controller to Ensure Performance and Temperature Constraints for Dynamic Applications

Noltsis

Zambelis

Catthoor

et al. 2019

ACM Trans. Embed. Comput. Syst.

View full text Add to dashboard Cite

To secure correct system operation, a plethora of Reliability, Availability and Serviceability (RAS) techniques have been deployed by circuit designers. RAS mechanisms however, come with the cost of extra clock cycles. In addition, a wide variety of dynamic workloads and different input conditions often constitute preemptive dependability techniques hard to implement. To this end, we focus on a realistic case study of a closed-loop controller that mitigates performance variation with a reactive response. This concept has been discussed but was only illustrated on small benchmarks. In particular, the extension of the approach to manage performance of dynamic workloads on a target platform has not been shown earlier. We compare our scheme against the version of a Linux CPU frequency governor in terms of timing response and energy consumption. Finally, we move forward and suggest a new flavor of our controller to efficiently manage processor temperature. Again, the concept is illustrated with a realistic case study and compared to a modern temperature manager.

show abstract

Classification of Resilience Techniques Against Functional Errors at Higher Abstraction Layers of Digital Systems

Cited by 5 publications

References 126 publications

Reliability Modeling and Mitigation for Embedded Memories

Reliability Modeling and Mitigation for Embedded Memories

Preliminaries

A Closed-Loop Controller to Ensure Performance and Temperature Constraints for Dynamic Applications

Contact Info

Product

Resources

About