# Cost-aware Duration Prediction for Software Upgrades in Datacenters

arXiv: https://arxiv.org/abs/2212.05155 · HTML: https://mlsanity.com/paper/2212.05155/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 47 references that are in the map.
- First posted on arXiv 10 Dec 2022.

## Builds on (most cited first, 5 of 5)

- [Learning Scheduling Algorithms for Data Processing Clusters](https://mlsanity.com/paper/1810.01963/) (2018)
- [An Alternative Probabilistic Interpretation of the Huber Loss](https://mlsanity.com/paper/1911.02088/) (2019)
- [The Case for Task Sampling based Learning for Cluster Job Scheduling](https://mlsanity.com/paper/2108.10464/) (2021)
- [NURD: Negative-Unlabeled Learning for Online Datacenter Straggler Prediction](https://mlsanity.com/paper/2203.08339/) (2022)
- [Don't cry over spilled records: Memory elasticity of data-parallel applications and its application to cluster scheduling](https://mlsanity.com/paper/1702.04323/) (2017)

## Abstract (arXiv)

Software upgrades are critical to maintaining server reliability in datacenters. While job duration prediction and scheduling have been extensively studied, the unique challenges posed by software upgrades remain largely under-explored. This paper presents the first in-depth investigation into software upgrade scheduling at datacenter scale. We begin by characterizing various types of upgrades and then frame the scheduling task as a constrained optimization problem. To address this problem, we introduce Acela, a cost-aware duration prediction framework designed to improve upgrade scheduling efficiency and throughput while meeting service-level objectives (SLOs). Acela accounts for asymmetric misprediction costs, strategically selects the best predictive models, and mitigates straggler-induced overestimations. Evaluations on Meta's production datacenter systems demonstrate that Acela significantly increases efficiency of the existing upgrade scheduler by improving upgrade window utilization by 1.25X, increasing the number of scheduled and completed upgrades by 33% and 41%, and reducing cancellation rates by 2.4X. The code and data sets will be released after paper acceptance.