The Online Pause and Resume Problem: Optimal Algorithms and An Application to Carbon-Aware Load Shifting

We introduce and study the online pause and resume problem. In this problem, a player attempts to find the $k$ lowest (alternatively, highest) prices in a sequence of fixed length $T$, which is revealed sequentially. At each time step, the player is presented with a price and decides whether to accept or reject it. The player incurs a switching cost whenever their decision changes in consecutive time steps, i.e., whenever they pause or resume purchasing. This online problem is motivated by the goal of carbon-aware load shifting, where a workload may be paused during periods of high carbon intensity and resumed during periods of low carbon intensity and incurs a cost when saving or restoring its state. It has strong connections to existing problems studied in the literature on online optimization, though it introduces unique technical challenges that prevent the direct application of existing algorithms. Extending prior work on threshold-based algorithms, we introduce double-threshold algorithms for both the minimization and maximization variants of this problem. We further show that the competitive ratios achieved by these algorithms are the best achievable by any deterministic online algorithm. Finally, we empirically validate our proposed algorithm through case studies on the application of carbon-aware load shifting using real carbon trace data and existing baseline algorithms.

[1]  John C.S. Lui,et al.  Pareto-Optimal Learning-Augmented Algorithms for Online k-Search Problems , 2022, ArXiv.

[2]  Sébastien Bubeck,et al.  The Randomized 𝑘-Server Conjecture Is False! , 2022, STOC.

[3]  Walid A. Hanafy,et al.  Ecovisor: A Virtual Energy System for Carbon-Efficient Applications , 2022, ASPLOS.

[4]  John C.S. Lui,et al.  The Online Knapsack Problem with Departures , 2022, Proc. ACM Meas. Anal. Comput. Syst..

[5]  Nicolas H. Christianson,et al.  Chasing Convex Bodies and Functions with Black-Box Advice , 2022, COLT.

[6]  Peng Li,et al.  Expert-Calibrated Learning for Online Optimization with Switching Costs , 2022, Proc. ACM Meas. Anal. Comput. Syst..

[7]  Nicolas H. Christianson,et al.  Smoothed Online Optimization with Unreliable Predictions , 2022, Proc. ACM Meas. Anal. Comput. Syst..

[8]  Fiodar Kazhamiaka,et al.  Carbon Explorer: A Holistic Framework for Designing Carbon Aware Datacenters , 2022, ASPLOS.

[9]  Donald F. Towsley,et al.  Competitive Algorithms for Online Multidimensional Knapsack Problems , 2021, Proc. ACM Meas. Anal. Comput. Syst..

[10]  Lauritz Thamsen,et al.  Let's wait awhile: how temporal workload shifting can reduce carbon emissions in the cloud , 2021, Middleware.

[11]  Adam Wierman,et al.  Pareto-Optimal Learning-Augmented Algorithms for Online Conversion Problems , 2021, NeurIPS.

[12]  Adam Wierman,et al.  Enabling Sustainable Clouds: The Case for Virtualizing the Energy System , 2021, SoCC.

[13]  Walfredo Cirne,et al.  Carbon-Aware Computing for Datacenters , 2021, IEEE Transactions on Power Systems.

[14]  Tongxin Li,et al.  Competitive Algorithms for the Online Multiple Knapsack Problem with Application to Electric Vehicle Charging , 2020, Proc. ACM Meas. Anal. Comput. Syst..

[15]  Prashant Shenoy,et al.  Hedge Your Bets: Optimizing Long-term Cloud Costs by Mixing VM Purchasing Options , 2020, 2020 IEEE International Conference on Cloud Engineering (IC2E).

[16]  Christian Coester,et al.  Online Metric Algorithms with Untrusted Predictions , 2020, ICML.

[17]  Prashant J. Shenoy,et al.  Combining Renewable Solar and Open Air Cooling for Greening Internet-Scale Distributed Networks , 2019, e-Energy.

[18]  Mark Sellke Chasing Convex Bodies Optimally , 2019, SODA.

[19]  Yin Tat Lee,et al.  Chasing Nested Convex Bodies Nearly Optimally , 2018, SODA.

[20]  James R. Lee,et al.  Metrical task systems on trees via mirror descent and unfair gluing , 2018, SODA.

[21]  Adam Wierman,et al.  Smoothed Online Convex Optimization in High Dimensions via Online Balanced Descent , 2018, COLT.

[22]  Zongpeng Li,et al.  Optimal Posted Prices for Online Cloud Resource Allocation , 2017, Proc. ACM Meas. Anal. Comput. Syst..

[23]  David E. Irwin,et al.  Transient Guarantees: Maximizing the Value of Idle Cloud Capacity , 2016, SC16: International Conference for High Performance Computing, Networking, Storage and Analysis.

[24]  Esther Mohr,et al.  Online algorithms for conversion problems: A survey , 2014 .

[25]  Adam Wierman,et al.  Renewable and cooling aware workload management for sustainable data centers , 2012, SIGMETRICS '12.

[26]  Lachlan L. H. Andrew,et al.  Online algorithms for geographical load balancing , 2012, 2012 International Green Computing Conference (IGCC).

[27]  Lachlan L. H. Andrew,et al.  Dynamic right-sizing for power-proportional data centers , 2011, 2011 Proceedings IEEE INFOCOM.

[28]  Deeparnab Chakrabarty,et al.  Budget constrained bidding in keyword auctions and online knapsack problems , 2008, WINE.

[29]  Konstantinos Panagiotou,et al.  Optimal Algorithms for k-Search with Application in Option Pricing , 2007, Algorithmica.

[30]  Ran El-Yaniv,et al.  Optimal Search and One-Way Trading Online Algorithms , 2001, Algorithmica.

[31]  Nathan Linial,et al.  On convex body chasing , 1993, Discret. Comput. Geom..

[32]  Allan Borodin,et al.  An optimal on-line algorithm for metrical task system , 1992, JACM.

[33]  Lachlan L. H. Andrew,et al.  Greening Geographical Load Balancing , 2015, IEEE/ACM Transactions on Networking.

[34]  S. Stewart ON CERTAIN INEQUALITIES INVOLVING THE LAMBERT W FUNCTION , 2009 .

[35]  A. Hoorfar,et al.  INEQUALITIES ON THE LAMBERTW FUNCTION AND HYPERPOWER FUNCTION , 2008 .