DROPS

Document

DOI: 10.4230/LIPIcs.ECRTS.2026.17

Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions

Authors: Abigail Eisenklam, Carlos A. Montenegro G., Xian Wang, Yifan Cai, Robert Gifford, Linh Thi Xuan Phan, and Ricardo G. Sanfelice

Published in: LIPIcs, Volume 375, 38th European Conference on Real-Time Systems (ECRTS 2026)

Abstract

Predictable timing on multicore systems requires careful management of shared resources such as the last-level cache and memory bandwidth. This paper presents MPORA, an uncertainty-aware dynamic resource allocation framework for multi-path, input-dependent real-time tasks on multicore platforms. MPORA models each job as a discrete-time dynamical system that captures execution dynamics and resource-dependent performance indicators. At runtime, MPORA monitors job execution states and predicts short-term instruction rates and remaining execution times under candidate allocations using predictive models trained offline. It then solves a receding-horizon optimization problem to compute resource allocations that maximize system-wide progress while meeting job deadlines. To address prediction uncertainty, MPORA integrates weighted conformal prediction into the optimization formulation, enabling uncertainty-aware deadline constraints. We implement MPORA as a Linux kernel module with microsecond-scale inference overhead. Experimental results on SPEC CPU benchmarks show that MPORA delivers accurate predictions under unseen inputs and distribution shifts with low overhead, while improving schedulability and response times over existing methods.

Cite as

Abigail Eisenklam, Carlos A. Montenegro G., Xian Wang, Yifan Cai, Robert Gifford, Linh Thi Xuan Phan, and Ricardo G. Sanfelice. Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions. In 38th European Conference on Real-Time Systems (ECRTS 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 375, pp. 17:1-17:26, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{eisenklam_et_al:LIPIcs.ECRTS.2026.17,
  author =	{Eisenklam, Abigail and Montenegro G., Carlos A. and Wang, Xian and Cai, Yifan and Gifford, Robert and Phan, Linh Thi Xuan and Sanfelice, Ricardo G.},
  title =	{{Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions}},
  booktitle =	{38th European Conference on Real-Time Systems (ECRTS 2026)},
  pages =	{17:1--17:26},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-429-1},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{375},
  editor =	{Kritikakou, Angeliki},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ECRTS.2026.17},
  URN =		{urn:nbn:de:0030-drops-266095},
  doi =		{10.4230/LIPIcs.ECRTS.2026.17},
  annote =	{Keywords: multicore, resource allocation, optimal control, learning, multi-path programs}
}

Document

Artifact

DOI: 10.4230/DARTS.12.2.9

Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions (Artifact)

Authors: Abigail Eisenklam, Carlos A. Montenegro G., Xian Wang, Yifan Cai, Robert Gifford, Linh Thi Xuan Phan, and Ricardo G. Sanfelice

Published in: DARTS, Volume 12, Issue 2, Special Issue of the 38th European Conference on Real-Time Systems (ECRTS 2026)

Abstract

This artifact accompanies the paper Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions appearing in ECRTS 2026. It is distributed as a Docker image and provides the scripts and datasets needed to reproduce the paper’s core empirical results across six SPEC CPU 2017 benchmarks. The artifact trains and evaluates gradient boosting regression trees (GBRT) that predict two fine-grained execution properties of each benchmark (remaining execution time and next-window instruction rate) under different resource allocations, exports each model to a compact Q16.16 fixed-point binary, evaluates the speed, accuracy, and size of the models using a fixed-point C inference library, and applies weighted conformal prediction (WCP) to produce high-probability upper bounds on the prediction errors under different operating scenarios. The in-kernel decision logic, which implements MPORA: Multi-Path Online Resource Allocation, is included as source for inspection only, as it requires a custom kernel and specialized hardware to run.

Cite as

Abigail Eisenklam, Carlos A. Montenegro G., Xian Wang, Yifan Cai, Robert Gifford, Linh Thi Xuan Phan, and Ricardo G. Sanfelice. Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions (Artifact). In Special Issue of the 38th European Conference on Real-Time Systems (ECRTS 2026). Dagstuhl Artifacts Series (DARTS), Volume 12, Issue 2, pp. 9:1-9:4, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@Article{eisenklam_et_al:DARTS.12.2.9,
  author =	{Eisenklam, Abigail and Montenegro G., Carlos A. and Wang, Xian and Cai, Yifan and Gifford, Robert and Phan, Linh Thi Xuan and Sanfelice, Ricardo G.},
  title =	{{Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions (Artifact)}},
  pages =	{9:1--9:4},
  journal =	{Dagstuhl Artifacts Series},
  ISSN =	{2509-8195},
  year =	{2026},
  volume =	{12},
  number =	{2},
  editor =	{Eisenklam, Abigail and Montenegro G., Carlos A. and Wang, Xian and Cai, Yifan and Gifford, Robert and Phan, Linh Thi Xuan and Sanfelice, Ricardo G.},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DARTS.12.2.9},
  URN =		{urn:nbn:de:0030-drops-266261},
  doi =		{10.4230/DARTS.12.2.9},
  annote =	{Keywords: multicore, resource allocation, optimal control, learning, multi-path programs}
}

Search Results

Documents authored by Wang, Xian

Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions

Abstract

Cite as

Uncertainty-Aware Resource Allocation for Multi-Path Programs with In-Kernel Predictions (Artifact)

Abstract

Cite as

Thanks for your feedback!

Could not send message