A case study of spark resource configuration and management for image processing applications
Bibliographic record
Abstract
The world population is expected to reach an estimated 9.8 billion by 2050, necessitating substantial increases in food production. Achieving such increases will require large-scale application of computer informatics within the agricultural sector. In particular, application of informatics to crop breeding has the potential to greatly enhance our ability to develop new varieties quickly and economically. Achieving this potential, however, will require capabilities for analyzing huge volumes of data acquired from various field-deployed image acquisition technologies. Although numerous frameworks for big data processing have been developed, there are relatively few published case studies that describe user experiences with these frameworks in particular application science domains. In this paper, we describe our efforts to apply Apache Spark to three applications of initial interest within the Plant Phenotyping and Imaging Research Centre (P2IRC) at the University of Saskatchewan. We find that default Spark parameter settings do not work well for these applications. We carry out extensive performance experiments to investigate the impact of alternative Spark parameter settings, both for applications run individually and in scenarios with multiple concurrently executing applications. We find that optimizing Spark parameter settings is challenging, but can yield substantial performance improvements, particularly with concurrent applications, provided that the dataset characteristics are considered. This is a first step towards insights regarding Spark parameter tuning on these classes of applications that may be more generally applicable to broader ranges of applications.
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 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.003 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".