ATOP-space and time adaptation for parallel and grid applications via flexible data partitioning
Bibliographic record
Abstract
Adaptive resource allocation is becoming an important feature to run parallel and grid applications: to better share space and time according to current workload, to schedule around obstacles as from reservation, to deal with varying system load under time-shared execution, and to deal with lack of accurate predictability on heterogeneous resources. Adaptation is potentially very expensive if total data repartitioning is required. Existing approaches of implementing large numbers of MPI via threads suffer from frequent thread switches, inefficient local communication, and being fixed to the chosen number of threads. Our ATOP middleware provides an approach which uses as many processes as there are processors and partitions and migrates the data, while processing the data per process as one data collection. For the partitioning and migration, we employ the Zoltan load-balancing library which is highly portable and supports a large variety of load-balancing approaches, including those of ParMETIS and Jostle. Exploiting features of Zoltan, we propose pre-partitioning (over-partitioning) of data graphs (reducing adaptation cost down to 25%) but can also flexibly decide to partition from scratch (for cases where over-partitioning does not perform well or where non-fitting numbers of resources need to be chosen).
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".