Speed-Oblivious Online Scheduling: Knowing (Precise) Speeds is not Necessary

Alexander Lindermayr, Nicole Megow, Martin Rapp
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:21312-21334, 2023.

Abstract

We consider online scheduling on unrelated (heterogeneous) machines in a speed-oblivious setting, where an algorithm is unaware of the exact job-dependent processing speeds. We show strong impossibility results for clairvoyant and non-clairvoyant algorithms and overcome them in models inspired by practical settings: (i) we provide competitive learning-augmented algorithms, assuming that (possibly erroneous) predictions on the speeds are given, and (ii) we provide competitive algorithms for the speed-ordered model, where a single global order of machines according to their unknown job-dependent speeds is known. We prove strong theoretical guarantees and evaluate our findings on a representative heterogeneous multi-core processor. These seem to be the first empirical results for scheduling algorithms with predictions that are evaluated in a non-synthetic hardware environment.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-lindermayr23a, title = {Speed-Oblivious Online Scheduling: Knowing ({P}recise) Speeds is not Necessary}, author = {Lindermayr, Alexander and Megow, Nicole and Rapp, Martin}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {21312--21334}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/lindermayr23a/lindermayr23a.pdf}, url = {https://proceedings.mlr.press/v202/lindermayr23a.html}, abstract = {We consider online scheduling on unrelated (heterogeneous) machines in a speed-oblivious setting, where an algorithm is unaware of the exact job-dependent processing speeds. We show strong impossibility results for clairvoyant and non-clairvoyant algorithms and overcome them in models inspired by practical settings: (i) we provide competitive learning-augmented algorithms, assuming that (possibly erroneous) predictions on the speeds are given, and (ii) we provide competitive algorithms for the speed-ordered model, where a single global order of machines according to their unknown job-dependent speeds is known. We prove strong theoretical guarantees and evaluate our findings on a representative heterogeneous multi-core processor. These seem to be the first empirical results for scheduling algorithms with predictions that are evaluated in a non-synthetic hardware environment.} }
Endnote
%0 Conference Paper %T Speed-Oblivious Online Scheduling: Knowing (Precise) Speeds is not Necessary %A Alexander Lindermayr %A Nicole Megow %A Martin Rapp %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-lindermayr23a %I PMLR %P 21312--21334 %U https://proceedings.mlr.press/v202/lindermayr23a.html %V 202 %X We consider online scheduling on unrelated (heterogeneous) machines in a speed-oblivious setting, where an algorithm is unaware of the exact job-dependent processing speeds. We show strong impossibility results for clairvoyant and non-clairvoyant algorithms and overcome them in models inspired by practical settings: (i) we provide competitive learning-augmented algorithms, assuming that (possibly erroneous) predictions on the speeds are given, and (ii) we provide competitive algorithms for the speed-ordered model, where a single global order of machines according to their unknown job-dependent speeds is known. We prove strong theoretical guarantees and evaluate our findings on a representative heterogeneous multi-core processor. These seem to be the first empirical results for scheduling algorithms with predictions that are evaluated in a non-synthetic hardware environment.
APA
Lindermayr, A., Megow, N. & Rapp, M.. (2023). Speed-Oblivious Online Scheduling: Knowing (Precise) Speeds is not Necessary. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:21312-21334 Available from https://proceedings.mlr.press/v202/lindermayr23a.html.

Related Material