Hardware-assisted fast routing for runtime reconfigurable computing
Bibliographic record
Abstract
Reconfigurable devices, such as the Field Programmable Gate Array, offer 10–100× computational density and reduced latency compared to conventional processor solutions. Despite its advantages, however, use of reconfigurable computing remains limited, largely due to the lack of software expressibility and longevity across various generations of devices. SCORE is a stream-based computational model that virtualizes reconfigurable computing resources by segmenting computations into fixed-size “pages” and time-multiplexing the virtual pages on available hardware. Therefore, SCORE applications can scale up or down automatically to operate seamlessly with a wide variety of hardware sizes. To support the SCORE model, we provide details of SCOREμP—a micro-architecture that combines (1) a reconfigurable array for regular fine-grained computation and (2) a sequential processor to run the page scheduler and execute SCORE operators or other user applications that do not run efficiently in spatial implementations. To fully realize the benefits of rapid partial reconfiguration of field-programmable devices, the runtime system often needs to schedule computing tasks dynamically and generate instance-specific configurations, i.e., new graphs which must be routed during program execution. Consequently, route time can be a significant overhead cost, reducing the achievable net benefits of dynamic configuration generation. By adding hardware to accelerate routing, it is possible to (1) compute routes in one one-thousandth of the time required by a traditional software router; and (2) achieve routes that are within five percent of state-of-the-art offline routing algorithms for a sample set of application netlists, and within three percent for the Toronto Place and Route Benchmarks. Strategic use of parallelism can allow total route time to scale substantially less than linearly in graph size. The observed speedups vary from greater than 10× with modest hardware overhead, to greater than 1000× with full hardware assistance.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
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".