StreamBucket: In-Network Adaptation for Late-Binding Stream Processing Systems
Bibliographic record
Abstract
Stream processing applications are increasingly de-ployed on hierarchical geo-distributed edge-cloud environments. Applications in such environments require frequent deployment reconfigurations to balance performance and cost by optimally utilizing the diverse resources. Recently, late-binding stream processing frameworks using a hierarchical network of routers have been introduced to seamlessly execute such reconfigurations with minimal impact on application performance. However, late-binding frameworks face scalability challenges as the routers can become a bottleneck. This restricts the maximum throughput that can be achieved by the network. We propose StreamBucket, a novel protocol that increases the capacity of late-binding routers by intelligently batching the tuples at various levels of the routing network and employs header compression techniques to reduce bandwidth. We design a model to predict the optimal batch size for unseen workloads, which reduces the amount of costly profiling. Our evaluations show that StreamBucket achieves up to 5x improvement in throughput using multi-tier batching and between 15% to 85% in bandwidth savings when using our protocol's compressed header.
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 | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.001 | 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".