Toward Adaptive Tracing: Efficient System Behavior Analysis using Language Models
Bibliographic record
Abstract
Tracing, a technique essential for unraveling the complexities of computer systems' behavior, involves the organized collection of low-level events, enabling anomaly identification, performance debugging, and root cause analysis. However, the significant overhead it imposes on large-scale systems, particularly in terms of performance and storage, has made it a less favorable tool for system maintenance. Previous efforts to mitigate tracing's burden have mostly centered around automating trace analysis but have primarily neglected the duration of events, a significant aspect of the information provided by tracers. To address these challenges, we propose an Adaptive Tracing method that leverages Language Models and kernel trace for precise system modeling. This novel approach minimizes overhead by recording detailed traces only during significant behavioral shifts and focusing on subsystems related to the root cause. Using a multi-task model, incorporating system call sequences and durations, we propose a root cause analysis method, enhancing model transparency and enabling targeted system tracing. Evaluation using a dataset of normal and noisy traces from an Apache server reveals that our Adaptive Tracer captures events related to abrupt changes with only 5.8% loss, reducing the collected trace by 77.1%, and accurately determining the respective noise set with 91.3% accuracy, outperforming previous state-of-the-art trace models by 20.9%.
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 | 0.014 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| 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".