SCinet DTN-as-a-Service Framework
Bibliographic record
Abstract
Transferring big data over Wide Area Networks (WANs) is challenging because optimization is dependent on the specifics of multiple parameters. Network services, paths, and technologies have different characteristics, including loss rate, latency, and available capacity. Yet, frameworks currently used to configure and orchestrate transfer systems, measure performance, and analyze results have limited capabilities. We propose a framework, DTN-as-a-Service (DaaS), for high-performance network data transfers using and integration of techniques, including virtualization, network provisioning, and performance data analysis. This framework has a modular design for supporting multiple transfer tools, optimizers and orchestrators for the data transfer environment, including \textit{Docker} and \textit{Kubernetes}. We present a \textit{Jupyter} based workflow for high-speed network data transfer in data-intensive science and evaluate the performance of the transfer with a simple programmable visualizer implemented in the framework. This framework has been implemented as a prototype at two recent SC supercomputing conferences. With the increase in the number and the capacity of WAN links at the conferences (multiple 100 Gbps WAN circuits), the challenges involved in setting up, testing, debugging, verifying and running applications on high-performance systems connecting to the conference SCinet WAN circuits also increase. The SCinet implementation of the DaaS framework for the conference community allowed users to control hardware, software, and network infrastructure for high-speed network data transfer, primarily for large scale applications. Through the evaluation of the framework in our test setup, we demonstrated that NVMe over Fabrics with TCP is twice as efficient compared to using conventional TCP in high-speed NVMe-to-NVMe transfers. We also implemented a 400 Gbps LAN experiment to evaluate the DaaS framework.
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 | 0.003 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.004 | 0.004 |
| Research integrity | 0.001 | 0.004 |
| Insufficient payload (model declined to judge) | 0.015 | 0.007 |
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".