The Impact of Hardware Variability on Applications Packaged with Docker and Guix: a Case Study in Neuroimaging
Bibliographic record
Abstract
The reproducibility of neuroimaging analyses across computational environments has gained significant attention over the last few years. While software containerization solutions such as Docker and Singularity have been deployed to mask the effects of software-induced variability, variations in hardware architectures still impact neuroimaging results in an unclear way. We study the effect of hardware variability on linear registration results produced by the FSL FLIRT application, a widely-used software component in neuroimaging data analyses. Using the Grid’5000 infrastructure, we study the effect of nine different CPU models using two software packaging systems (Docker and Guix), and we compare the resulting hardware variability to numerical variability measured with random rounding. Results show that hardware, software, and numerical variability lead to perturbations of similar magnitudes — albeit uncorrelated — suggesting that these three types of variability act as independent sources of numerical noise with similar magnitude. Therefore, random rounding is a practical solution to measure the effect of numerical noise induced by hardware variability in this application. The effect of hardware perturbations on linear registration remains moderate, with average translation errors of 0.1 mm (maximum: 0.5 mm) and average rotation errors of 0.02 deg (maximum: 0.2 deg). Such variations might impact downstream analyses when linear registration is used as initialization step for other operations.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.045 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".