|
In this thesis we present contributions in the following areas:
1.5.1 Quantifying and Characterising Cloud Performance
We have made several contributions to the area of Cloud performance measurement, particularly in respect to variation amongst instances of the same type, helping to: (1) quantify it; (2) identify causes and extent; and (3) characterize it. Some of this work has already been cited in a major survey of performance literature by Leitner and Cito (2016), who evaluated 54 papers contributing to empirical evaluation of Public Infrastructure Clouds. Leitner and Cito’s citations included:
-
John O’Loughlin and Lee Gillam (2014) "Performance Evaluation for Cost-Efficient Public Infrastructure Cloud Use". In J. Altmann et al. (Eds.): International Conference on Economics of Grids, Clouds, Systems and Services (GECON) 2014, LNCS 8914, pp. 133–145, 2014.
-
John O’Loughlin and Lee Gillam (2013) "Performance Prediction for Public Infrastructure Clouds: an EC2 Case Study". 2013 IEEE International Conference on Cloud Computing Technology and Science (CloudCom 2013)
-
Lee Gillam, Bin Li, John O’Loughlin and Anuz Pratap Singh Tomar (2013) "Fair Benchmarking for Cloud Computing Systems". Springer Open Journal of Cloud Computing 2:6. DOI: 10.1186/2192-113X-2-6
1.5.2 A Consistent Performance Model of Instance Performance
A consistent model of instance performance must be capable of generating performance data for a group of instances with the following properties: (1) over time it produces qualitatively similar time-plots to those empirically observed and (2) a cross-section at any given moment in time must have the same characteristics as cross-sectional variations empirically observed. In this thesis we present a consistent model of instance performance, which, as far as we aware, is a first. A model satisfying property (2) was presented in:
-
John O’Loughlin and Lee Gillam (2017) “A Performance Brokerage for Heterogeneous Clouds”. Future Generation Computer Systems (FGCS). In press. https://doi.org/10.1016/j.future.2017.05.005
1.5.3 Evaluation of Risk in Performance Improvement Strategies
Farley et al. (2012) and Ou et al. (2013) have proposed performance improvement strategies, with an objective of obtaining instances with particular performance levels or CPU models. We re-appraise this work with respect to the level of risk, and we find assumptions typically made regarding allocation of CPUs models to instances are not supported by empirical evidence. Further, through simulation, we demonstrate how these assumptions lead to an underestimation of the risk i.e. the standard deviation in the expected costs. Such strategies, then, are, shown to be demonstrably riskier than previously thought.
1.5.4 A Profitable Cloud Broker
To be sustainable, a Cloud broker must create a value-add that users are prepared to pay for. We demonstrate that the performance broker as proposed may be profitable under certain conditions currently found on Clouds. We also show how the broker may be sensitive to market competition, vagaries in demand, and gaming strategies the clients may be able to employ. As far as we are aware the only other work to have demonstrated a profitable broker is Rogers and Cliff (2012). The work presented in this thesis extends that presented by the publication listed below by: (1) use of a consistent performance model; and (2) introduction of a mechanism for client/broker negotiation.
-
John O’Loughlin and Lee Gillam (2017) “A Performance Brokerage for Heterogeneous Clouds”. Future Generation Computer Systems (FGCS). In press. https://doi.org/10.1016/j.future.2017.05.005
1.5.5 A Co-operative Host Separation Detection Strategy
Instances are said to be co-locating if they are on the same physical host, otherwise they are host separated. We present a strategy which two instances can co-operatively employ, enabling them to positively determine if they are host separated, whilst a failure to show host separation suggests co-location and repeated failure more strongly hints at this. The strategy is suitable for Xen based Clouds, and is addressed by the two publications listed below.
-
John O'Loughlin and Lee Gillam (2016) "Sibling Virtual Machine Co-location Confirmation and Avoidance Tactics for Public Infrastructure Clouds". Journal of Supercomputing Volume 72, Issue 3, pp 961-984: DOI 10.1007/s11227-016-1627-9
-
John O'Loughlin and Lee Gillam (2015) "Addressing Issues of Cloud Resilience, Security and Performance through Simple Detection of Co-locating Sibling Virtual Machine Instances". 5th International Conference on Cloud Computing and Services Science (CLOSER 2015). [Short listed for best paper].
1.5.6 Publications
The work presented in this thesis has been published in a number of international peer reviewed journals and conferences, as well an EPSRC report on the cost of Cloud for research. Further, it has also been presented at a leading industry event, Cloud World Forum 2015 and Large Scale Complex IT Systems (LSCITS) Cloud Workshop, Grasmere, 2013 and 2014. In particular:
Published Papers:
-
John O’Loughlin and Lee Gillam (2017) “A Performance Brokerage for Heterogeneous Clouds”. Future Generation Computer Systems (FGCS). In Press. https://doi.org/10.1016/j.future.2017.05.005.
-
John O'Loughlin and Lee Gillam (2016) "Sibling Virtual Machine Co-location Confirmation and Avoidance Tactics for Public Infrastructure Clouds". Journal of Supercomputing Volume 72, Issue 3, pp 961-984: DOI 10.1007/s11227-016-1627-9
-
John O’Loughlin and Lee Gillam (2014) "Should Infrastructure Clouds be Priced Entirely on Performance? An EC2 Case Study". International Journal of Big Data Intelligence
-
John O'Loughlin and Lee Gillam (2015) "Addressing Issues of Cloud Resilience, Security and Performance through Simple Detection of Co-locating Sibling Virtual Machine Instances". 5th International Conference on Cloud Computing and Services Science (CLOSER 2015).
-
John O’Loughlin and Lee Gillam (2015) "Re-Appraising Instance Seeking in Public Clouds". IEEE Technically Co-Sponsored Science and Information Conference 2015, London UK.
-
John O’Loughlin and Lee Gillam (2014) "Performance Evaluation for Cost-Efficient Public Infrastructure Cloud Use". In J. Altmann et al. (Eds.): International Conference on Economics of Grids, Clouds, Systems and Services (GECON) 2014, LNCS 8914, pp. 133–145, 2014.
-
John O’Loughlin and Lee Gillam (2014) "Performance Prediction for Unseen Virtual Machines". 4th International Conference on Cloud Computing and Services Science (CLOSER 2014)
-
John O’Loughlin and Lee Gillam (2014) "Good Performance Metrics for Cloud Service Brokers". 5th International Conference on Cloud Computing, Grids and Virtualisation (CLOUD COMPUTING 2014)
-
John O’Loughlin and Lee Gillam (2013) "Performance Prediction for Public Infrastructure Clouds: an EC2 Case Study". 2013 IEEE International Conference on Cloud Computing Technology and Science (CloudCom 2013)
-
Lee Gillam, Bin Li, John O’Loughlin (2012) "Adding Cloud Performance to Service Level Agreements". 2nd International Conference on Cloud Computing and Services Science, CLOSER 2012
-
Lee Gillam, Bin Li and John O’Loughlin (2014) "Benchmarking cloud performance for service level agreement parameters". Int. J. Cloud Computing, Vol. 3, No. 1, pp.3–23
-
Lee Gillam, Bin Li, John O’Loughlin and Anuz Pratap Singh Tomar (2013) "Fair Benchmarking for Cloud Computing Systems". Springer Open Journal of Cloud Computing 2:6. DOI: 10.1186/2192-113X-2-6
-
Lee Gillam, Bin Li, John O’Loughlin (2013) "Benchmarking Cloud Performance for Service Level Agreement Parameters". International Journal of Cloud Computing, CloudSecGov Special Issue 2013.
-
Lee Gillam, Bin Li, John O’Loughlin (2012) "Teaching Clouds: Lessons Taught and Lessons Learnt". In Chao, L. (ed.) Cloud Computing for Teaching and Learning: Strategies for Design and Implementation
-
Bin Li, Lee Gillam and John O’Loughlin (2010) "Towards Application-Specific Service Level Agreements: Experiments in Clouds and Grids". In Antonopoulos and Gillam (Eds.), Cloud Computing: Principles, Systems and Applications. Springer-Verlag
-
Geoff Curtis, John O’Loughlin, Lee Gillam and Alex Curtis (2014) “Refresh of the Cost Analysis of Cloud Computing for Research”, Final Report to EPSRC and JISC
Dostları ilə paylaş: |
