Temporal Rate Limiting

Cloud elasticity at a flat fee

John S. Otto, Rade Stanojevic, Nikolaos Laoutaris

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

In the current usage-based pricing scheme offered by most cloud computing providers, customers are charged based on the capacity and the lease time of the resources they capture (bandwidth, number of virtual machines, IOPS rate, etc.). Taking advantage of this pricing scheme, customers can implement auto-scaling purchase policies by leasing (e.g., hourly) necessary amounts of resources to satisfy a desired QoS threshold under their current demand. Auto-scaling yields strict QoS and variable charges. Some customers, however, would be willing to settle for a more relaxed statistical QoS in exchange for a predictable flat charge. In this work we propose Temporal Rate Limiting (TRL), a purchase policy that permits a customer to allocate optimally a specified purchase budget over a predefined period of time. TRL offers the same expected QoS with auto-scaling but at a lower, flat charge. It also outperforms in terms of QoS a naive flat charge policy that splits the available budget uniformly in time. We quantify the benefits of TRL analytically and also deploy TRL on Amazon EC2 and perform a live validation in the context of a "blacklisting" application for Twitter.

Original languageEnglish
Title of host publication2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012
Pages151-156
Number of pages6
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012 - Orlando, FL, United States
Duration: 25 Mar 201230 Mar 2012

Other

Other2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012
CountryUnited States
CityOrlando, FL
Period25/3/1230/3/12

Fingerprint

Elasticity
Quality of service
Cloud computing
Costs
Bandwidth

ASJC Scopus subject areas

  • Computer Science(all)
  • Electrical and Electronic Engineering

Cite this

Otto, J. S., Stanojevic, R., & Laoutaris, N. (2012). Temporal Rate Limiting: Cloud elasticity at a flat fee. In 2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012 (pp. 151-156). [6193478] https://doi.org/10.1109/INFCOMW.2012.6193478

Temporal Rate Limiting : Cloud elasticity at a flat fee. / Otto, John S.; Stanojevic, Rade; Laoutaris, Nikolaos.

2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012. 2012. p. 151-156 6193478.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Otto, JS, Stanojevic, R & Laoutaris, N 2012, Temporal Rate Limiting: Cloud elasticity at a flat fee. in 2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012., 6193478, pp. 151-156, 2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012, Orlando, FL, United States, 25/3/12. https://doi.org/10.1109/INFCOMW.2012.6193478
Otto JS, Stanojevic R, Laoutaris N. Temporal Rate Limiting: Cloud elasticity at a flat fee. In 2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012. 2012. p. 151-156. 6193478 https://doi.org/10.1109/INFCOMW.2012.6193478
Otto, John S. ; Stanojevic, Rade ; Laoutaris, Nikolaos. / Temporal Rate Limiting : Cloud elasticity at a flat fee. 2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012. 2012. pp. 151-156
@inproceedings{59b75deb8a0e4a508ac8b7d5806a929f,
title = "Temporal Rate Limiting: Cloud elasticity at a flat fee",
abstract = "In the current usage-based pricing scheme offered by most cloud computing providers, customers are charged based on the capacity and the lease time of the resources they capture (bandwidth, number of virtual machines, IOPS rate, etc.). Taking advantage of this pricing scheme, customers can implement auto-scaling purchase policies by leasing (e.g., hourly) necessary amounts of resources to satisfy a desired QoS threshold under their current demand. Auto-scaling yields strict QoS and variable charges. Some customers, however, would be willing to settle for a more relaxed statistical QoS in exchange for a predictable flat charge. In this work we propose Temporal Rate Limiting (TRL), a purchase policy that permits a customer to allocate optimally a specified purchase budget over a predefined period of time. TRL offers the same expected QoS with auto-scaling but at a lower, flat charge. It also outperforms in terms of QoS a naive flat charge policy that splits the available budget uniformly in time. We quantify the benefits of TRL analytically and also deploy TRL on Amazon EC2 and perform a live validation in the context of a {"}blacklisting{"} application for Twitter.",
author = "Otto, {John S.} and Rade Stanojevic and Nikolaos Laoutaris",
year = "2012",
doi = "10.1109/INFCOMW.2012.6193478",
language = "English",
isbn = "9781467310178",
pages = "151--156",
booktitle = "2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012",

}

TY - GEN

T1 - Temporal Rate Limiting

T2 - Cloud elasticity at a flat fee

AU - Otto, John S.

AU - Stanojevic, Rade

AU - Laoutaris, Nikolaos

PY - 2012

Y1 - 2012

N2 - In the current usage-based pricing scheme offered by most cloud computing providers, customers are charged based on the capacity and the lease time of the resources they capture (bandwidth, number of virtual machines, IOPS rate, etc.). Taking advantage of this pricing scheme, customers can implement auto-scaling purchase policies by leasing (e.g., hourly) necessary amounts of resources to satisfy a desired QoS threshold under their current demand. Auto-scaling yields strict QoS and variable charges. Some customers, however, would be willing to settle for a more relaxed statistical QoS in exchange for a predictable flat charge. In this work we propose Temporal Rate Limiting (TRL), a purchase policy that permits a customer to allocate optimally a specified purchase budget over a predefined period of time. TRL offers the same expected QoS with auto-scaling but at a lower, flat charge. It also outperforms in terms of QoS a naive flat charge policy that splits the available budget uniformly in time. We quantify the benefits of TRL analytically and also deploy TRL on Amazon EC2 and perform a live validation in the context of a "blacklisting" application for Twitter.

AB - In the current usage-based pricing scheme offered by most cloud computing providers, customers are charged based on the capacity and the lease time of the resources they capture (bandwidth, number of virtual machines, IOPS rate, etc.). Taking advantage of this pricing scheme, customers can implement auto-scaling purchase policies by leasing (e.g., hourly) necessary amounts of resources to satisfy a desired QoS threshold under their current demand. Auto-scaling yields strict QoS and variable charges. Some customers, however, would be willing to settle for a more relaxed statistical QoS in exchange for a predictable flat charge. In this work we propose Temporal Rate Limiting (TRL), a purchase policy that permits a customer to allocate optimally a specified purchase budget over a predefined period of time. TRL offers the same expected QoS with auto-scaling but at a lower, flat charge. It also outperforms in terms of QoS a naive flat charge policy that splits the available budget uniformly in time. We quantify the benefits of TRL analytically and also deploy TRL on Amazon EC2 and perform a live validation in the context of a "blacklisting" application for Twitter.

UR - http://www.scopus.com/inward/record.url?scp=84862080731&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862080731&partnerID=8YFLogxK

U2 - 10.1109/INFCOMW.2012.6193478

DO - 10.1109/INFCOMW.2012.6193478

M3 - Conference contribution

SN - 9781467310178

SP - 151

EP - 156

BT - 2012 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2012

ER -