Abstract
In the cloud market, there exist multiple cloud providers adopting auction-based mechanisms to offer cloud resources to users. These auction-based cloud providers need to compete against each other to maximize the profits by setting the cloud resource prices effectively. In this paper, we analyze how an auction-based cloud provider sets the auction price effectively when competing against other cloud providers in the evolutionary market where the amount of participated cloud users is changing. The pricing strategy is affected by many factors, such as the auction price of its opponents, the prices charged to users in the previous round, the bidding behavior of cloud users, and so on. Therefore, we model this problem as a partially observable Markov game and adopt a gradient-based multi-agent deep reinforcement learning algorithm to generate the competing pricing strategy. We also run extensive experiments to evaluate our pricing strategy against other five benchmark pricing strategies in the auction-based cloud market. The experimental results show that our generated pricing strategy can beat other pricing strategies in terms of long-term profits and the amount of participated users, and it can also learn cloud users’ marginal values and their choice of cloud providers effectively.
Similar content being viewed by others
Notes
Note that our setting can be easily extended to the case with more than two competing cloud providers. In the real world, although there exist more than two cloud providers, users’ final choices are usually made in the two providers, and thus we focus on the competition between two cloud providers.
When cloud users choose the cloud provider, they have not actually participate and submit bids, and thus the cloud provider cannot determine the actual auction price at this stage. Therefore, we consider that cloud users use the average auction price as the expected payment of per unit resource when computing the expected utility.
References
Alzhouri F, Agarwal A, Liu Y (2021) Maximizing cloud revenue using dynamic pricing of multiple class virtual machines. IEEE Trans Cloud Comput 9(2):682–695
Åström KJ, Murray RM (2010) Feedback systems: an introduction for scientists and engineers. Princeton University Press, Princeton
Bichler M, Kalagnanam J, Katircioglu K et al (2002) Applications of flexible pricing in business-to-business electronic commerce. IBM Syst J 41:287–302
Cai Z, Li X, Ruiz R et al (2018) Price forecasting for spot instances in cloud computing. Future Gener Comput Syst 79:38–53
Cardellini V, Valerio VD, Presti FL (2020) Game-theoretic resource pricing and provisioning strategies in cloud systems. IEEE Trans Serv Comput 13(1):86–98
Dawoud W, Takouna I, Meinel C (2012) Reliable approach to sell the spare capacity in the cloud. In: Cloud computing, pp 229–236
Du B, Wu C, Huang Z (2019) Learning resource allocation and pricing for cloud profit maximization. AAAI Press, Palo Alto, pp 7570–7577
Feng Y, Li B, Li B (2013) Price competition in an oligopoly market with multiple IAAS cloud providers. IEEE Trans Comput 63:59–73
Jung H, Klein CM (2001) Optimal inventory policies under decreasing cost functions via geometric programming. Eur J Oper Res 132:628–642
Kansal S, Kumar H, Kaushal S et al (2018) Genetic algorithm-based cost minimization pricing model for on-demand IAAS cloud service. J Supercomput 76:1–26
Khandelwal V, Gupta CP, Chaturvedi AK (2018) Perceptive bidding strategy for amazon EC2 spot instance market. Multiagent Grid Syst 14:83–102
Kumar D, Baranwal G, Raza Z et al (2018) A survey on spot pricing in cloud computing. J Netw Syst Manag 26:809–856
Littman ML (1994) Markov games as a framework for multi-agent reinforcement learning. In: Proceedings of the eleventh international conference on machine learning (ML-94), pp 157–163
Lowe R, Wu Y, Tamar A, et al (2017) Multi-agent actor-critic for mixed cooperative-competitive environments. In: Advances in neural information processing systems, pp 6379–6390
Marschak J (1959) Binary choice constraints on random utility indicators. Cowles Foundation Discussion Papers
Mei J, Li K, Tong Z et al (2018) Profit maximization for cloud brokers in cloud computing. IEEE Trans Parallel Distrib Syst 30(1):190–203
Nguyen ND, Nguyen T, Nahavandi S (2019) Multi-agent behavioral control system using deep reinforcement learning. Neurocomputing 359:58–68
Pearl R, Reed LJ (1920) On the rate of growth of the population of the united states since 1790 and its mathematical representation. Proc Natl Acad Sci USA 6:275
Pinto L, Davidson J, Sukthankar R, et al (2017) Robust adversarial reinforcement learning. In: Proceedings of the 34th international conference on machine learning, pp 2817–2826
Rong J, Qin T, An B (2019) Competitive cloud pricing for long-term revenue maximization. J Comput Sci Technol 34(3):645–656
Shi B, Zhu H, Wang J, et al (2017) Optimize pricing policy in evolutionary market with multiple proactive competing cloud providers. In: 2017 IEEE 29th international conference on tools with artificial intelligence (ICTAI), pp 202–209
Silver D, van Hasselt H, Hessel M, et al (2017) The predictron: end-to-end learning and planning. In: Proceedings of the 34th international conference on machine learning, pp 3191–3199
Silver D, Schrittwieser J, Simonyan K et al (2017) Mastering the game of go without human knowledge. Nature 550:354–359
Train KE (2003) Discrete choice methods with simulation. Cambridge University Press, Cambridge
Truong-Huu T, Tham CK (2013) A game-theoretic model for dynamic pricing and competition among cloud providers. In: 2013 IEEE/ACM 6th international conference on utility and cloud computing, pp 235–238
Truong-Huu T, Tham CK (2014) A novel model for competition and cooperation among cloud providers. IEEE Trans Cloud Comput 2(3):251–265
Venugopal S (2009) Cloud computing and emerging it platforms: vision, hype, and reality for delivering computing as the 5th utility. Future Gener Comput Syst 25:599–616
Wang P, Qi Y, Hui D, et al (2013) Present or future: optimal pricing for spot instances. In: 2013 IEEE 33rd international conference on distributed computing systems, pp 410–419
Wu C, Toosi AN, Buyya R et al (2021) Hedonic pricing of cloud computing services. IEEE Trans Cloud Comput 9(1):182–196
Xu B, Qin T, Qiu G, et al (2015) Optimal pricing for the competitive and evolutionary cloud market. In: Twenty-fourth international joint conference on artificial intelligence, pp 139–145
Zhang F, Li J, Li Z (2020) A TD3-based multi-agent deep reinforcement learning method in mixed cooperation–competition environment. Neurocomputing 411:206–215
Zheng L, Joe-Wong C, Tan CW et al (2015) How to bid the cloud. ACM SIGCOMM Comput Commun Rev 45:71–84
Acknowledgements
This paper was funded by the Humanity and Social Science Youth Research Foundation of Ministry of Education (Grant No. 19YJC790111), the Philosophy and Social Science Post-Foundation of Ministry of Education (Grant No. 18JHQ060) and Shenzhen Fundamental Research Program (Grant No. JCYJ20190809175613332).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
In this section, we describe how to derive the probability of the cloud user choosing a cloud provider given its expected utility. Specifically, the probability of cloud user j choosing to be served by provider i at stage t is denoted as \(P_{j, i}^{t}\):
From Eq. 20, we can see that cloud user j will choose to be served by provider i only if the user’s utility is maximized. According to Eq. 7, we then get an expression for the choice probability:
Since \(\eta _{j,i}\) is independent, the cumulative distribution over all \(i^{\prime }\ne i\) is the product of the individual cumulative distributions:
Now, the choice probability is the integral of \(P_{j,i}^{t}\vert \eta _{j,i}\) over all values of \(\eta _{j,i}\) weighted by its density:
The closed-form expression is
which is the probability of user j choosing provider i at stage t.
Rights and permissions
About this article
Cite this article
Shi, B., Huang, L. & Shi, R. A deep reinforcement learning-based approach for pricing in the competing auction-based cloud market. SOCA 16, 83–95 (2022). https://doi.org/10.1007/s11761-022-00334-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11761-022-00334-8