Northern University Bangladesh
Track submission Sign in
NUB Journal of Science & Engineering e-ISSN 2789-5678 · p-ISSN 2789-1234
Original Research Computer Science & Engineering

Energy-Aware Task Scheduling for Edge Computing in Intermittent-Power Environments

  • Rehana Parvin ✉
  • Ayesha Siddika iD
  • Tanvir Ahmed iD
  1. 1. University of Dhaka
  2. 2. Northern University Bangladesh
  3. 3. Bangladesh University of Engineering and Technology
Received:
11 Jan 2026
Accepted:
6 Feb 2026
Published:
6 Mar 2026
Licence:
cc-by-4.0

Abstract

Edge deployments in areas with unreliable grid supply must schedule work around power availability rather than assuming it. We formulate energy-aware task scheduling under stochastic power interruption and propose a checkpointing policy that bounds expected work loss. Against a 14-month power-availability trace from rural Bangladesh, the policy completed 34% more tasks than energy-oblivious scheduling.

Keywords: distributed systems internet of things optimisation cloud computing

Full text

1. Introduction

The problem addressed in this article has attracted sustained attention over the past decade, yet a number of practical obstacles remain unresolved. This work sets out an approach that addresses those obstacles directly and evaluates it against established benchmarks.

2. Related Work

Prior work falls broadly into two strands. The first treats the problem as one of representation; the second as one of optimisation. Our contribution sits between the two, borrowing the representational insight of the former while retaining the tractability of the latter.

3. Methodology

Our approach proceeds in three stages: preprocessing, model construction and evaluation. Each is described in sufficient detail for independent reproduction, and our implementation is available on request.

4. Results

Across all evaluation conditions the proposed method performs at least as well as the strongest baseline, and substantially better in the low-data regime that motivated this work. Performance degrades gracefully as noise increases, which we attribute to the regularisation strategy described in section 3.2.

5. Limitations

Two limitations warrant emphasis. The evaluation is confined to a single geographic context, and generalisation beyond it is untested. The method also assumes a data volume that smaller institutions may not have.

6. Conclusion

We have presented and evaluated an approach to Computer Science & Engineering problems that improves on established baselines while remaining computationally tractable. Future work will extend the evaluation to additional contexts.

Declarations

Data availability
Data are available on reasonable request
Conflicts of interest
The authors declare no competing interests.
Acknowledgements
The authors thank the Faculty of Engineering for computing resources.