Learning predictive autoscaling policies for cloud-hosted microservices using trace-driven modeling
Author | Abdullah M. |
Author | Iqbal W. |
Author | Erradi A. |
Author | Bukhari F. |
Available date | 2020-04-01T06:50:40Z |
Publication Date | 2019 |
Publication Name | Proceedings of the International Conference on Cloud Computing Technology and Science, CloudCom |
Publication Name | 11th IEEE International Conference on Cloud Computing Technology and Science, CloudCom 2019, 19th IEEE International Conference on Computer and Information Technology, CIT 2019, 2019 International Workshop on Resource Brokering with Blockchain, RBchain 2019 and 2019 Asia-Pacific Services Computing Conference, APSCC 2019 |
Resource | Scopus |
ISBN | 978-1-7281-5011-6 |
ISBN | 978-1-7281-5012-3 |
ISSN | 23302194 |
Abstract | Autoscaling methods are important to ensure response time guarantees for cloud-hosted microservices. Most of the existing state-of-the-art autoscaling methods use rule-based reactive policies with static thresholds defined either on monitored resource consumption metrics such as CPU and memory utilization or application-level metrics such as the response time. However, it is challenging to determine the most appropriate threshold values to minimize resource consumption and performance violations. Whereas, predictive autoscaling methods can help to address these challenges. These methods require considerable time to collect sufficient performance traces representing different resource provisioning possibilities for a target infrastructure to train a useful predictive autoscaling model. In this paper, we tackle this problem by proposing a system that models the response time of microservices through stress testing and then uses a trace-driven simulation to learn a predictive autoscaling model for satisfying response time requirements automatically. The proposed solution reduces the need for collecting performance traces to learn a predictive autoscaling model. Our experimental evaluation on AWS cloud using a microservice under realistic dynamic workloads validates the proposed solution. The validation results show excellent performance to satisfy the response time requirement with only 4.5% extra cost for using the proposed autoscaling method compared to the reactive autoscaling method. - 2019 IEEE. |
Sponsor | Currently, we are extending our work by enabling feedback to the predictive autoscaling model for online retraining using operational performance traces. We also plan to automatically identify the best forecasting method and window sizes for fine tuning the proposed system. ACKNOWLEDGEMENT This publication was made possible by NPRP grant # NPRP9-224-1-049 from the Qatar National Research Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of the authors. |
Language | en |
Publisher | IEEE Computer Society |
Subject | Cloud computing Microservices Predictive autoscaling Response time guarantees SLO violations |
Type | Conference Paper |
Pagination | 119-126 |
Volume Number | 2019-December |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
-
Computer Science & Engineering [2288 items ]