AAPA: An A rchetype- A ware P redictive A utoscaler with Uncertainty Quantification for Serverless Workloads on Kubernetes 4 Corresponding author. 5 Equal contribution.
Guilin Zhang 1 2 5 , Srinivas Vippagunta 2 5 , Raghavendra Nandagopal 2 5 , Suchitra Raman 2 5 , Jeff Xu 2 5 , Marcus Pfeiffer 2 5 , Shreeshankar Chatterjee 2 5 , Ziqi Tan 1 5 , Wulan Guo 1 5 , Hailong Jiang 3 4 5 · 2025
Read AAPA: An A rchetype- A ware P redictive A utoscaler with Uncertainty Quantification for Serverless Workloads on Kubernetes 4 Corresponding author. 5 Equal contribution. by Guilin Zhang 1 2 5 , Srinivas Vippagunta 2 5 , Raghavendra Nandagopal 2 5 , Suchitra Raman 2 5 , Jeff Xu 2 5 , Marcus Pfeiffer 2 5 , Shreeshankar Chatterjee 2 5 , Ziqi Tan 1 5 , Wulan Guo 1 5 , Hailong Jiang 3 4 5 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-05653.txt.
Description
Guilin Zhang and co-authors present AAPA in 2025, a tool that scales serverless computing work on Kubernetes. Such platforms face wild swings in load, and plain rules react too late or too evenly. Past tools also ignored what kind of load was coming and how sure the forecast was.
AAPA sorts workloads into four shapes: spikes, cycles, ramps, and steady flows. Each shape gets its own scaling plan, tuned by forecast confidence. The team also released a labeled data set of 300,000 workload windows from Azure. AAPA cut rule breaks by up to half and delays by forty percent, though at higher cost under spiky loads.
Catalog record
| CREATOR | Guilin Zhang 1 2 5 , Srinivas Vippagunta 2 5 , Raghavendra Nandagopal 2 5 , Suchitra Raman 2 5 , Jeff Xu 2 5 , Marcus Pfeiffer 2 5 , Shreeshankar Chatterjee 2 5 , Ziqi Tan 1 5 , Wulan Guo 1 5 , Hailong Jiang 3 4 5 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-05653