Towards Performance Modeling of Speculative Execution for Cloud Applications
(2020) 3rd Workshop on Hot Topics in Cloud Computing Performance- Abstract
- Interesting approaches to counteract performance variability within cloud datacenters include sending multiple request clones, either immediately or after a specified waiting time. In this paper we present a performance model of cloud applications that utilize the latter concept, known as speculative execution. We study the popular Join-Shortest-Queue load-balancing strategy under the processor sharing queuing discipline. Utilizing the near-synchronized service property of this setting, we model speculative execution using a simplified synchronized service model. Our model is approximate, but accurate enough to be useful even for high utilization scenarios. Furthermore, the model is valid for any, possibly empirical, inter-arrival and... (More)
- Interesting approaches to counteract performance variability within cloud datacenters include sending multiple request clones, either immediately or after a specified waiting time. In this paper we present a performance model of cloud applications that utilize the latter concept, known as speculative execution. We study the popular Join-Shortest-Queue load-balancing strategy under the processor sharing queuing discipline. Utilizing the near-synchronized service property of this setting, we model speculative execution using a simplified synchronized service model. Our model is approximate, but accurate enough to be useful even for high utilization scenarios. Furthermore, the model is valid for any, possibly empirical, inter-arrival and service time distributions. We present preliminary simulation results, showing the promise of our proposed model. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/f7840eb4-293b-4a92-a034-7ae1c8b82d52
- author
- Nylander, Tommi LU ; Ruuskanen, Johan LU ; Årzén, Karl-Erik LU and Maggio, Martina LU
- organization
- publishing date
- 2020
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Cloning, Speculative Execution, Cloud Computing, Datacenters
- host publication
- ACM/SPEC International Conference on Performance Engineering Companion (ICPE ’20 Companion)
- pages
- 3 pages
- conference name
- 3rd Workshop on Hot Topics in Cloud Computing Performance
- conference location
- Edmonton, Canada
- conference dates
- 2020-04-21 - 2020-04-21
- external identifiers
-
- scopus:85086043228
- DOI
- 10.1145/3375555.3384379
- project
- Control-based resource management in the distributed cloud
- Event-Based Information Fusion for the Self-Adaptive Cloud
- WASP: Autonomous Cloud
- language
- English
- LU publication?
- yes
- id
- f7840eb4-293b-4a92-a034-7ae1c8b82d52
- date added to LUP
- 2020-02-25 18:10:26
- date last changed
- 2022-04-18 20:57:54
@inproceedings{f7840eb4-293b-4a92-a034-7ae1c8b82d52, abstract = {{Interesting approaches to counteract performance variability within cloud datacenters include sending multiple request clones, either immediately or after a specified waiting time. In this paper we present a performance model of cloud applications that utilize the latter concept, known as speculative execution. We study the popular Join-Shortest-Queue load-balancing strategy under the processor sharing queuing discipline. Utilizing the near-synchronized service property of this setting, we model speculative execution using a simplified synchronized service model. Our model is approximate, but accurate enough to be useful even for high utilization scenarios. Furthermore, the model is valid for any, possibly empirical, inter-arrival and service time distributions. We present preliminary simulation results, showing the promise of our proposed model.}}, author = {{Nylander, Tommi and Ruuskanen, Johan and Årzén, Karl-Erik and Maggio, Martina}}, booktitle = {{ACM/SPEC International Conference on Performance Engineering Companion (ICPE ’20 Companion)}}, keywords = {{Cloning; Speculative Execution; Cloud Computing; Datacenters}}, language = {{eng}}, title = {{Towards Performance Modeling of Speculative Execution for Cloud Applications}}, url = {{https://lup.lub.lu.se/search/files/76541762/HotCloudPerf_2020_tommi_final.pdf}}, doi = {{10.1145/3375555.3384379}}, year = {{2020}}, }