Automated Bitstream-Level Cost-Reliability Design-Space Exploration for SRAM-Based FPGAs
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Triple Modular Redundancy (TMR) is a common approach to mitigate the effects of Single-Event Upsets (SEUs) in SRAM-based Field-Programmable Gate Arrays (FPGAs), where these faults may cause changes in the configuration of logic or interconnect resources. Partial TMR aims at balancing SEU mitigation with redundancy costs. This work introduces a Design-Space Exploration (DSE) approach that automatically generates and evaluates cost-reliability-optimized, Pareto-optimal partial TMR configurations of modules in a hierarchical design. The approach is evaluated using a proof-of-concept implementation for AMD’s 7 Series FPGAs and five case-study designs, including the NEORV32 RISC-V CPU. Multiple fitness assignment variants – based on static bitstream analysis, (statistical) fault injection results, and a combined approach –-are compared regarding effectiveness and runtime. Comparing the hypervolumes of the generated Pareto fronts of the final generation and a randomly generated starting generation, the approach improves cost-effectiveness of the generated TMR solutions by 17%–52%, delivering an attractive benefit-cost-ratio. The presented approach effectively generates a diverse set of TMR solutions across a wide cost-reliability range, allowing the designer to choose a variant that best fulfills the application’s, mission’s, or mission phase’s cost-reliability requirements.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it