SpotOn: a batch computing service for the spot market

@inproceedings{Subramanya2015SpotOnAB,
  title={SpotOn: a batch computing service for the spot market},
  author={Supreeth Subramanya and Tian Guo and Prateek Sharma and David E. Irwin and Prashant J. Shenoy},
  booktitle={SoCC},
  year={2015}
}
Cloud spot markets enable users to bid for compute resources, such that the cloud platform may revoke them if the market price rises too high. Due to their increased risk, revocable resources in the spot market are often significantly cheaper (by as much as 10×) than the equivalent non-revocable on-demand resources. One way to mitigate spot market risk is to use various fault-tolerance mechanisms, such as checkpointing or replication, to limit the work lost on revocation. However, the… CONTINUE READING

Figures, Results, and Topics from this paper.

Key Quantitative Results

  • Our simulation results using a job trace from a Google cluster indicate that SpotOn lowers costs by 91.9% compared to using on-demand resources with little impact on performance.
  • Our simulation results using a job trace from a Google cluster indicate that SpotOn low­ers costs by 91.9% compared to using on-demand resources with little impact on performance.
  • Our results on a Google cluster trace indicate that SpotOn lowers costs by 91.9% compared to using on-demand re­sources with little impact on performance.
  • When compared to a cost-aware policy that only uses checkpointing, SpotOn reduces cost by up to 74%, again while increasing job completion time by only 2%.

Similar Papers

Citations

Publications citing this paper.
SHOWING 1-10 OF 50 CITATIONS

System Support for Managing Risk in Cloud Computing Platforms

VIEW 10 EXCERPTS
CITES BACKGROUND, METHODS & RESULTS
HIGHLY INFLUENCED

Transiency-driven Resource Management for Cloud Computing Platforms

VIEW 10 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

The Financialization of Cloud Computing: Opportunities and Challenges

  • 2017 26th International Conference on Computer Communication and Networks (ICCCN)
  • 2017
VIEW 7 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

SpotLight: An Information Service for the Cloud

  • 2016 IEEE 36th International Conference on Distributed Computing Systems (ICDCS)
  • 2016
VIEW 4 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Cost-Effective Cloud Server Provisioning for Predictable Performance of Big Data Analytics

  • IEEE Transactions on Parallel and Distributed Systems
  • 2019
VIEW 4 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

The Price is ( Not ) Right : Reflections on Pricing for Transient Cloud Servers

David Irwin, Prashant Shenoy, +3 authors Ahmed Ali-Eldin
  • 2019
VIEW 5 EXCERPTS
CITES BACKGROUND & RESULTS
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2016
2019

CITATION STATISTICS

  • 12 Highly Influenced Citations

  • Averaged 13 Citations per year from 2017 through 2019

References

Publications referenced by this paper.
SHOWING 1-2 OF 2 REFERENCES

Reliable Provisioning of Spot Instances for Compute-intensive Applications

  • 2012 IEEE 26th International Conference on Advanced Information Networking and Applications
  • 2011
VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

Google Clusterusage Traces: Format + Schema

C. Reiss, J. Wilkes, J. L. Hellerstein
  • Technical report, Google Inc.,
  • 2011
VIEW 3 EXCERPTS
HIGHLY INFLUENTIAL