research-article

HotSpot: automated server hopping in cloud spot markets

Authors:

Supreeth Shastri,

David IrwinAuthors Info & Claims

SoCC '17: Proceedings of the 2017 Symposium on Cloud Computing

Pages 493 - 505

https://doi.org/10.1145/3127479.3132017

Published: 24 September 2017 Publication History

Abstract

Cloud spot markets offer virtual machines (VMs) for a dynamic price that is much lower than the fixed price of on-demand VMs. In exchange, spot VMs expose applications to multiple forms of risk, including price risk, or the risk that a VM's price will increase relative to others. Since spot prices vary continuously across hundreds of different types of VMs, flexible applications can mitigate price risk by moving to the VM that currently offers the lowest cost. To enable this flexibility, we present HotSpot, a resource container that "hops" VMs---by dynamically selecting and self-migrating to new VMs---as spot prices change. HotSpot containers define a migration policy that lowers cost by determining when to hop VMs based on the transaction costs (from vacating a VM early and briefly double paying for it) and benefits (the expected cost savings). As a side effect of migrating to minimize cost, HotSpot is also able to reduce the number of revocations without degrading performance. HotSpot is simple and transparent: since it operates at the systems-level on each host VM, users need only run an HotSpot-enabled VM image to use it. We implement a HotSpot prototype on EC2, and evaluate it using job traces from a production Google cluster. We then compare HotSpot to using on-demand VMs and spot VMs (with and without fault-tolerance) in EC2, and show that it is able to lower cost and reduce the number of revocations without degrading performance.

References

[1]

2017. Amazon EC2 Pricing, https://aws.amazon.com/ec2/pricing/on-demand/. (Accessed August 2017).

[2]

2017. Amazon Spot Instances, https://aws.amazon.com/ec2/spot/. (Accessed August 2017).

[3]

2017. AWS Global Infrastructure, https://aws.amazon.com/about-aws/global-infrastructure/. (Accessed August 2017).

[4]

2017. cmpute.io, Inc. http://www.cmpute.io. (Accessed August 2017).

[5]

2017. Google Compute Engine, https://cloud.google.com/compute/pricing. (Accessed August 2017).

[6]

2017. Linux Containers, http://linuxcontainers.org. (Accessed August 2017).

[7]

2017. Spot Instance Limits. http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-spot-limits.html. (Accessed August 2017).

[8]

2017. Spot Instance Product Details, https://aws.amazon.com/ec2/spot/details/. (Accessed August 2017).

[9]

C. Babcock. 2015. Amazon's 'Virtual CPU'? You Figure it Out, In Information Week. (December 23rd 2015).

[10]

G. Banga, P. Druschel, and J. Mogul. 1999. Resource Containers: A New Facility for Resource Management in Server Systems. In Symposium on Operating Systems Design and Implementation (OSDI).

[11]

J. Barr. 2015. New - EC2 Spot Instance Termination Notices. AWS Blog, https://aws.amazon.com/blogs/aws/new-ec2-spot-instance-termination-notices/. (January 6th 2015).

[12]

J. Barr. 2015. New Spot Fleet Option - Distribute Your Fleet Across Multiple Capacity Pools. AWS Blog, https://aws.amazon.com/blogs/aws/new-spot-fleet-option-distribute-your-fleet-across-multiple-capacity-pools/. (September 15th 2015).

[13]

M. Ben-Yehuda, M. Day, Z. Dubitzky, M. Factor, N. Har'El, A. Gordon, A. Liguori, O. Wasserman, and B. Yassour. 2010. The Turtles Project: Design and Implementation of Nested Virtualization. In Symposium on Operating Systems Design and Implementation (OSDI).

[14]

O. Ben-Yehuda, M. Ben-Yehuda, A. Schuster, and D. Tsafrir. 2011. Deconstructing Amazon EC2 Spot Instance Pricing. In International Conference on Cloud Computing Technology and Science (CloudCom).

[15]

O. Ben-Yehuda, M. Ben-Yehuda, A. Schuster, and D. Tsafrir. 2013. Deconstructing Amazon EC2 Spot Instance Pricing. ACM Transactions on Economics and Computation (TEAC) 1, 3 (2013).

Digital Library

[16]

J. Boutelle. 2011. What to do when Amazon's spot prices spike, In Gigaom. (December 27th 2011).

[17]

L. Chaufournier, P. Sharma, P. Shenoy, and Y. Tay. 2016. Containers and Virtual Machines at Scale: A Comparative Study. In International Middleware Conference (Middleware).

[18]

C. Clark, K. Fraser, S. Hand, J. Hansen, E. Jul, C. Limpach, I. Pratt, and A. Waeldr. 2005. Live Migration of Virtual Machines. In Symposium on Networked Systems Design and Implementation (NSDI).

[19]

L. Columbus. 2016. Roundup of Cloud Computing Forecasts andMarketEstimar.es, In Forbes. (March 13th 2016).

[20]

A. Harlap, A. Tumanov, A. Chung, G. Ganger, and P. Gibbons. 2017. Proteus: Agile ML Elasticity through Tiered Reliability in Dynamic Resource Markets. In European Conference on Computer Systems (EuroSys).

[21]

S. Higginbotham. 2013. Bidding Strategies? Arbitrage? AWS Spot Market is where Computing and Finance Meet, In Gigaom. (October 8th 2013).

[22]

B. Huang, N. Jarrett, S. Babu, S. Mukherjee, and J. Yang. 2015. Cumulon: Matrix-Based Data Analytics in the Cloud with Spot Instances. Proceedings of the VLDB Endowment (PVLDB) 9, 3 (November 2015).

Digital Library

[23]

Q. Jia, Z. Shen, W. Song, R. van Renesse, and H. Weatherspoon. 2016. Smart Spot Instances for the Supercloud. In International Workshop on CrossCloud Infrastructures and Platforms (CrossCloud).

[24]

F. Lardinois. 2016. Spotinst, which helps you buy AWS Spot Instances, raises $2M Series A, In TechCrunch. (March 8th 2016).

[25]

Q. Liang, C. Wang, and B. Urgaonkar. 2016. Spot Characterization: What are the Right Features to Model?. In International Workshop on System Analytics and Characterization (SAC).

[26]

M. Mazzucco and M. Dumas. 2011. Achieving Performance and Availability Guarantees with Spot Instances. In International Conference on High Performance Computing and Communications (HPCC).

[27]

I. Menache, O. Shamir, and N. Jain. 2014. On-demand, Spot, or Both: Dynamic Resource Allocation for Executing Batch Jobs in the Cloud. In International Conference on Autonomic Computing (ICAC).

[28]

J. Novet. 2015. Amazon pays $20M--$50M for ClusterK, the startup that can run apps on AWS at 10% of the regular price, In VentureBeat. (April 29th 2015).

[29]

X. Ouyang, D. Irwin, and P. Shenoy. 2016. Spotlight: An Information Service for the Cloud. In International Conference on Distributed Computing Systems (ICDCS).

[30]

Charles Reiss, John Wilkes, and Joseph L. Hellerstein. 2011. Google cluster-usage traces: format + schema. Technical Report. Google Inc., Mountain View, CA, USA. Revised 2014-11-17 for version 2.1. Posted at https://github.com/google/cluster-data.

[31]

P. Sharma, T. Guo, X. He, D. Irwin, and P. Shenoy. 2016. Flint: Batch-Interactive Data-Intensive Processing on Transient Servers. In European Conference on Computer Systems (EuroSys).

[32]

P. Sharma, D. Irwin, and P. Shenoy. 2016. How Not to Bid the Cloud. In Workshop on Hot Topics in Cloud Computing (HotCloud).

[33]

P. Sharma, D. Irwin, and P. Shenoy. 2017. Portfolio-driven Resource Management for Transient Cloud Servers. In International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS).

[34]

P. Sharma, S. Lee, T. Guo, D. Irwin, and P. Shenoy. 2015. SpotCheck: Designing a Derivative IaaS Cloud on the Spot Market. In European Conference on Computer Systems (EuroSys).

[35]

S. Shastri and D. Irwin. 2017. Towards Index-based Global Trading in Cloud Markets. In Workshop on Hot Topics in Cloud Computing (HotCloud).

[36]

Z. Shen, Q. Jia, E. Sela, B. Rainero, W. Song, R. van Renesse, and H. Weatherspoon. 2016. Follow the Sun through the Clouds: Application Migration for Geographically Shifting Workloads. In Symposium on Cloud Computing (SoCC).

[37]

R. Singh, D. Irwin, P. Shenoy, and K.K. Ramakrishnan. 2013. Yank: Enabling Green Data Centers to Pull the Plug. In Symposium on Networked Systems Design and Implementation (NSDI).

Digital Library

[38]

R. Singh, P. Sharma, D. Irwin, P. Shenoy, and K.K. Ramakrishnan. 2014. Here Today, Gone Tomorrow: Exploiting Transient Servers in Datacenters. IEEE Internet Computing 18, 4 (April 2014).

[39]

Y. Song, M. Zafer, and K. Lee. 2012. Optimal Bidding in Spot Instance Market. In Infocom.

[40]

S. Subramanya, T. Guo, P. Sharma, D. Irwin, and P. Shenoy. 2015. SpotOn: A Batch Computing Service for the Spot Market. In Symposium on Cloud Computing (SoCC).

[41]

S. Tang, J. Yuan, and X. Li. 2012. Towards Optimal Bidding Strategy for Amazon EC2 Cloud Spot Instance. In International Conference on Cloud Computing (CLOUD).

[42]

John Wilkes. 2011. More Google cluster data. Google research blog. (Nov. 2011). Posted at http://googleresearch.blogspot.com/2011/ll/more-google-cluster-data.html.

[43]

D. Williams, H. Jamjoom, and H. Weatherspoon. 2012. The Xen-Blanket: Virtualize Once, Run Everywhere. In European Conference on Computer Systems (EuroSys).

[44]

R. Wolski and J. Brevik. 2016. Providing Statistical Reliability Guarantees in the AWS Spot Tier. In High Performance Computing Symposium (HPC).

[45]

T. Wood, A. Lagar-Cavilla, K. K. Ramakrishnan, P. Shenoy, and J. Van der Merwe. 2011. PipeCloud: Using Causality to Overcome Speed-of-Light Delays in Cloud-Based Disaster Recovery. In Symposium on Cloud Computing (SoCC).

[46]

T. Wood, K. K. Ramakrishnan, P. Shenoy, and J. Van der Merwe. 2011. CloudNet: Dynamic Pooling of Cloud Resources by Live WAN Migration of Virtual Machines. In International Conference on Virtual Execution Environments (VEE).

[47]

Z. Xu, C. Stewart, N. Deng, and X. Wang. 2016. Blending On-Demand and Spot Instances to Lower Costs for In-Memory Storage. In International Conference on Computer Communications (Infocom).

[48]

Y. Yan, Y. Gao, Z. Guo, B. Chen, and T. Moscibroda. 2016. TR-Spark: Transient Computing for Big Data Analytics. In Symposium on Cloud Computing (SoCC).

[49]

Y. Yang, G. Kim, W. Song, Y. Lee, A. Chung, Z. Qian, B. Cho, and B. Chun. 2017. Pado: A Data Processing Engine for Harnessing Transient Resources in Datacenters. In European Conference on Computer Systems (EuroSys).

[50]

M. Zafer, Y. Song, and K. Lee. 2012. Optimal Bids for Spot VMs in a Cloud for Deadline Constrained Jobs. In International Conference on Cloud Computing (CLOUD).

[51]

S. Zaman and D. Grosu. 2011. Efficient Bidding for Virtual Machine Instances in Clouds. In International Conference on Cloud Computing (CLOUD).

[52]

Q. Zhang, E. Gürses, R. Boutaba, and J. Xiao. 2011. Dynamic Resource Allocation for Spot Markets in Clouds. In Workshop on Hot Topics in Management of Internet, CLoud, and Enterprise Networks and Services (HotICE).

[53]

L. Zheng, C. Joe-Wong, C. Tan, M. Chiang, and X. Wang. 2015. How to Bid the Cloud. In ACM SIGCOMM Conference (SIGCOMM).

Cited By

Mandal SLal SSensarma SSaha SMaji GKhatua SDas R(2024)An Online Algorithm for Cost Minimization of Amazon EC2 Burstable ResourcesDistributed Computing and Intelligent Technology10.1007/978-3-031-50583-6_8(117-132)Online publication date: 4-Jan-2024
https://doi.org/10.1007/978-3-031-50583-6_8
Park HGanger GAmvrosiadis GGilad YKostic DMoatti YBiran O(2023)Mimir: Finding Cost-efficient Storage Configurations in the Public CloudProceedings of the 16th ACM International Conference on Systems and Storage10.1145/3579370.3594776(22-34)Online publication date: 5-Jun-2023
https://dl.acm.org/doi/10.1145/3579370.3594776
Mandal SMaji GKhatua SDas R(2023)Cost Minimizing Reservation and Scheduling Algorithms for Public CloudsIEEE Transactions on Cloud Computing10.1109/TCC.2021.313346411:2(1365-1380)Online publication date: 1-Apr-2023
https://doi.org/10.1109/TCC.2021.3133464
Show More Cited By

Index Terms

HotSpot: automated server hopping in cloud spot markets
1. Computer systems organization
  1. Architectures
    1. Distributed architectures
      1. Cloud computing

Recommendations

Stratus: cost-aware container scheduling in the public cloud
SoCC '18: Proceedings of the ACM Symposium on Cloud Computing

Stratus is a new cluster scheduler specialized for orchestrating batch job execution on virtual clusters, dynamically allocated collections of virtual machine instances on public IaaS platforms. Unlike schedulers for conventional clusters, Stratus ...
Cloud Index Tracking: Enabling Predictable Costs in Cloud Spot Markets
SoCC '18: Proceedings of the ACM Symposium on Cloud Computing

Cloud spot markets rent VMs for a variable price that is typically much lower than the price of on-demand VMs, which makes them attractive for a wide range of large-scale applications. However, applications that run on spot VMs suffer from cost ...
Dynamic Cloud Pricing for Revenue Maximization

In cloud computing, a provider leases its computing resources in the form of virtual machines to users, and a price is charged for the period they are used. Though static pricing is the dominant pricing strategy in today's market, intuitively price ...

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

SoCC '17: Proceedings of the 2017 Symposium on Cloud Computing

September 2017

672 pages

ISBN:9781450350280

DOI:10.1145/3127479

Copyright © 2017 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 24 September 2017

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Conference

SoCC '17

Sponsor:

SoCC '17: ACM Symposium on Cloud Computing

September 24 - 27, 2017

California, Santa Clara

Acceptance Rates

Overall Acceptance Rate 169 of 722 submissions, 23%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

41
Total Citations
View Citations
604
Total Downloads

Downloads (Last 12 months)46
Downloads (Last 6 weeks)11

Reflects downloads up to

Other Metrics

View Author Metrics

Citations

Cited By

Mandal SLal SSensarma SSaha SMaji GKhatua SDas R(2024)An Online Algorithm for Cost Minimization of Amazon EC2 Burstable ResourcesDistributed Computing and Intelligent Technology10.1007/978-3-031-50583-6_8(117-132)Online publication date: 4-Jan-2024
https://doi.org/10.1007/978-3-031-50583-6_8
Park HGanger GAmvrosiadis GGilad YKostic DMoatti YBiran O(2023)Mimir: Finding Cost-efficient Storage Configurations in the Public CloudProceedings of the 16th ACM International Conference on Systems and Storage10.1145/3579370.3594776(22-34)Online publication date: 5-Jun-2023
https://dl.acm.org/doi/10.1145/3579370.3594776
Mandal SMaji GKhatua SDas R(2023)Cost Minimizing Reservation and Scheduling Algorithms for Public CloudsIEEE Transactions on Cloud Computing10.1109/TCC.2021.313346411:2(1365-1380)Online publication date: 1-Apr-2023
https://doi.org/10.1109/TCC.2021.3133464
Alourani AKshemkalyani A(2022)Cost-Effective Spot Instances Provisioning Using Features of Cloud MarketsInternational Journal of Cloud Applications and Computing10.4018/IJCAC.30827612:1(1-27)Online publication date: 30-Nov-2022
https://doi.org/10.4018/IJCAC.308276
Iqbal SLi HBergsma SBeschastnikh IHu AGavrilovska AAltınbüken DBinnig C(2022)CoSpotProceedings of the 13th Symposium on Cloud Computing10.1145/3542929.3563499(540-556)Online publication date: 7-Nov-2022
https://dl.acm.org/doi/10.1145/3542929.3563499
Kadupitiya JJadhao VSharma P(2022)SciSpot: Scientific Computing On Temporally Constrained Cloud Preemptible VMsIEEE Transactions on Parallel and Distributed Systems10.1109/TPDS.2022.315727233:12(3575-3588)Online publication date: 1-Dec-2022
https://dl.acm.org/doi/10.1109/TPDS.2022.3157272
Lin LPan LLiu S(2022)A Cost-Effective Framework for Running Industrial Big Data Analysis Applications in Public CloudsIEEE Internet of Things Journal10.1109/JIOT.2021.31221969:13(10554-10562)Online publication date: 1-Jul-2022
https://doi.org/10.1109/JIOT.2021.3122196
Wang ZXiao WZhou BHan J(2022)Qos-Sure: Qos-Assurance AutoScaling of Sharing GPU For DNN Inference in Multi-Task2022 10th International Conference on Information Systems and Computing Technology (ISCTech)10.1109/ISCTech58360.2022.00064(367-372)Online publication date: Dec-2022
https://doi.org/10.1109/ISCTech58360.2022.00064
Wong WCorneo LZavodovski AKangasharju J(2022)PISTIS: Composing AWS Spot Instances with GuaranteesICC 2022 - IEEE International Conference on Communications10.1109/ICC45855.2022.9838290(5469-5474)Online publication date: 16-May-2022
https://doi.org/10.1109/ICC45855.2022.9838290
Lin LPan LLiu S(2022)Learning to make auto-scaling decisions with heterogeneous spot and on-demand instances via reinforcement learningInformation Sciences: an International Journal10.1016/j.ins.2022.10.071614:C(480-496)Online publication date: 1-Oct-2022
https://dl.acm.org/doi/10.1016/j.ins.2022.10.071
Show More Cited By

View Options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Media

Figures

Other

Tables

View Table of Contents