SPA: Harnessing Availability in the AWS Spot Market

Amazon Web Services (AWS) offers transient virtual servers at a discounted price as a way to sell unused spare capacity in its data centers. Although transient servers are very appealing as some instances have up to 90% discount, they are not bound to regular availability guarantees as they are opportunistic resources sold on the spot market. In this paper, we present SPA, a framework that remarkably increases the spot instance reliability over time due to insights gained from the analysis of historical data, such as cross-region price variability and intervals between evictions. We implemented the SPA reliability strategy, evaluated them using over one year of historical pricing data from AWS, and found out that we can increase the transient instance lifetime by adding a pricing overhead of 3.5% in the spot price in the best scenario.

[1]  David E. Irwin,et al.  HotSpot: automated server hopping in cloud spot markets , 2017, SoCC.

[2]  Artur Andrzejak,et al.  Monetary Cost-Aware Checkpointing and Migration on Amazon Cloud Spot Instances , 2012, IEEE Transactions on Services Computing.

[3]  Shaojie Tang,et al.  Towards Optimal Bidding Strategy for Amazon EC2 Cloud Spot Instance , 2012, 2012 IEEE Fifth International Conference on Cloud Computing.

[4]  Luís E. T. Rodrigues,et al.  Hourglass: Leveraging Transient Resources for Time-Constrained Graph Processing in the Cloud , 2019, EuroSys.

[5]  Robbert van Renesse,et al.  Economics of a supercloud , 2016, CrossCloud@EuroSys.

[6]  Prashant Dahiwale,et al.  An efficient dynamic resource allocation strategy for VM environment in cloud , 2015, 2015 International Conference on Pervasive Computing (ICPC).

[7]  Lukas Häfner,et al.  Scheduling Flexible Demand in Cloud Computing Spot Markets , 2019, Business & Information Systems Engineering.

[8]  Ian A. Kash,et al.  Fixed and market pricing for cloud services , 2012, 2012 Proceedings IEEE INFOCOM Workshops.

[9]  Muli Ben-Yehuda,et al.  Deconstructing Amazon EC2 Spot Instance Pricing , 2011, 2011 IEEE Third International Conference on Cloud Computing Technology and Science.

[10]  Xin He,et al.  Flint: batch-interactive data-intensive processing on transient servers , 2016, EuroSys.

[11]  Yang Chen,et al.  TR-Spark: Transient Computing for Big Data Analytics , 2016, SoCC.

[12]  Mayutan Arumaithurai,et al.  OFM: An Online Fisher Market for Cloud Computing , 2019, IEEE INFOCOM 2019 - IEEE Conference on Computer Communications.

[13]  Prateek Sharma,et al.  SpotCheck: designing a derivative IaaS cloud on the spot market , 2015, EuroSys.

[14]  Jussi Kangasharju,et al.  Bricklayer: Resource Composition on the Spot Market , 2020, ICC 2020 - 2020 IEEE International Conference on Communications (ICC).

[15]  Prateek Sharma,et al.  SpotOn: a batch computing service for the spot market , 2015, SoCC.

[16]  Gregory R. Ganger,et al.  Tributary: spot-dancing for elastic services with latency SLOs , 2018, USENIX ATC.

[17]  Robbert van Renesse,et al.  Smart spot instances for the supercloud , 2016, CrossCloud@EuroSys.

[18]  Mikhail Khodak,et al.  Learning Cloud Dynamics to Optimize Spot Instance Bidding Strategies , 2018, IEEE INFOCOM 2018 - IEEE Conference on Computer Communications.

[19]  Sven Seuken,et al.  Cloud Pricing: The Spot Market Strikes Back , 2019, EC.

[20]  Liang Zheng,et al.  How to Bid the Cloud , 2015, Comput. Commun. Rev..

[21]  Jussi Kangasharju,et al.  Container deployment strategy for edge networking , 2019 .