An Empirical Evaluation of the Viability of the Serverless Paradigm for Scientific Workflows
Bibliographic record
Abstract
Scientific workflows are typically data-intensive. They consist of many stages, each of which may contain hundreds to even thousands of tasks. Traditionally, scientific workflows have been executed using the serverful computing model. Serverless computing presents an attractive alternative to the serverful computing model as it frees developers from managing and provisioning resources and offers a fine-grained billing model. In this work, we study the viability and evaluate the trade-offs of using the serverless paradigm to run scientific workflows. We follow an empirical approach to evaluate the performance and cost benefits of this paradigm and to study the suitability of the current serverless software stack to support complex data-intensive scientific workflows. Specifically, we discuss, implement, and evaluate three orchestration approaches for executing scientific workflows: serverful-centralized, serverless-centralized, and serverless-decentralized. This work is the first to implement and evaluate a purely serverless orchestration approach that does not require deploying a dedicated workflow manager for scientific workflows. Our evaluation shows that serverless orchestration approaches cause a noticeable performance overhead for some workflow patterns (e.g., reduce stages) due to accessing a large amount of remote data. We propose two optimizations (i.e., prefetching file privileges and container placement) that exploit data locality to mitigate that impact. Our evaluation with three scientific workflows—Montage, 1000Genomes, and SRA Search—shows that serverless-centralized and serverless-decentralized achieve a comparable performance to a serverful approach. Also, our results show that prefetching file privileges and container placement optimizations improve the performance by 32.6% and 44% respectively when compared to an unoptimized version for the Montage application. We also introduce a cost model to estimate which approach is cheaper for a given application and a cloud provider. Our cost analysis shows that answering this question depends on the characteristics of the workflow and the pricing of the cloud provider.
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 | 0.029 | 0.099 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.001 | 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".