MODE: mix driven on-line resource demand estimation
Bibliographic record
Abstract
Adaptive performance management solutions often rely on models that require accurate resource demand measures that are estimated in an on-line manner. However it is typically not possible to directly measure resource demands at the abstraction they are needed, e.g., for a software service within an application server that is invoked by a URL. For such cases, linear regression techniques are often used to estimate resource demands. We evaluate the effectiveness of the Least Squares (LSQ) and Least Absolute Deviations (LAD) regression techniques, used extensively by others, as well as Support Vector Regression (SVR) for the purpose of demand estimation. To the best of our knowledge SVR has not yet been evaluated for computer resource demand estimation. We consider the predictive accuracy of these methods for three different real and simulated workloads. Our results demonstrate the importance of tuning the regression parameters of the techniques. We propose an on-line method named Mix Driven On-line Resource Demand Estimation (MODE) that automatically and quickly tunes the regression parameters for LSQ, LAD, and SVR to achieve their best results. The method is novel in that it relies on pre-defined workload mixes with known aggregate demand values to support the tuning exercise. We show that when employed in an on-line manner, tuning with respect to pre-defined mixes is significantly more accurate than the traditional approach of using only step by step data.
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 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".