ABSTRACT
In an effort to increase the utilization of data center resources cloud providers have introduced a new type of virtual machine (VM) offering, called a burstable VM (BVM). Our work is the first to study the characteristics of burstable VMs (based on traces from production systems at a major cloud provider) and resource allocation approaches for BVM workloads. We propose new approaches for BVM resource allocation and use extensive simulations driven by field data to compare them with two baseline approaches used in practice. We find that traditional approaches based on using a fixed oversubscription ratio or based on the Central Limit Theorem do not work well for BVMs: They lead to either low utilization or high server capacity violation rates. Based on the lessons learned from our workload study, we develop a new approach to BVM scheduling, called Audible, using a non-parametric statistical model, which makes the approach light-weight and workload independent, and obviates the need for training machine learning models and for tuning their parameters. We show that Audible achieves high system utilization while being able to enforce stringent requirements on server capacity violations.
- Ahsan Ali, Riccardo Pinciroli, Feng Yan, and Evgenia Smirni. CEDULE: A scheduling framework for burstable performance in cloud computing. In IEEE International Conference on Autonomic Computing (ICAC), pages 141--150. IEEE, 2018.Google ScholarCross Ref
- Ahsan Ali, Riccardo Pinciroli, Feng Yan, and Evgenia Smirni. It's not a sprint, it's a marathon: Stretching multi-resource burstable performance in public clouds. In Dejan S. Milojicic and Vinod Muthusamy, editors, Proceedings of the 20th International Middleware Conference Industrial Track, pages 36--42. ACM, 2019.Google ScholarDigital Library
- Ataollah Fatahi Baarzi, Timothy Zhu, and Bhuvan Urgaonkar. Burscale: Using burstable instances for cost-effective autoscaling in the public cloud. In Proceedings of the ACM Symposium on Cloud Computing, SoCC 2019, Santa Cruz, CA, USA, November 20-23, 2019, pages 126--138. ACM, 2019.Google Scholar
- Noman Bashir, Nan Deng, Krzysztof Rzadca, David Irwin, Sree Kodak, and Rohit Jnagal. Take it to the limit: peak prediction-driven resource overcommitment in datacenters. In Proceedings of the Sixteenth European Conference on Computer Systems (EuroSys), pages 556--573, 2021.Google ScholarDigital Library
- R.N. Calheiros, E. Masoumi, R. Ranjan, and R. Buyya. Workload prediction using arima model and its impact on cloud applications' QoS. IEEE Transactions on Cloud Computing, 2014.Google Scholar
- Maxime C. Cohen, Philipp Keller, Vahab Mirrokni, and Morteza Zadimoghaddam. Overcommitment in cloud services - bin packing with chance constraints. Management Science, 2019.Google Scholar
- Eli Cortez, Anand Bonde, Alexandre Muzio, Mark Russinovich, Marcus Fontoura, and Ricardo Bianchini. Resource central: Understanding and predicting workloads for improved resource management in large cloud platforms. In Proceedings of the 26th Symposium on Operating Systems Principles (SoSP), pages 153--167. ACM, 2017.Google ScholarDigital Library
- Nan Deng, Zichen Xu, Christopher Stewart, and Xiaorui Wang. Telltale tails: Decomposing response times for live internet services. In Sixth International Green and Sustainable Computing Conference, IGSC, pages 1--8. IEEE Computer Society, 2015.Google Scholar
- Rick Durrett. Probability: Theory and Examples. Cambridge, 4 edition, 2010.Google ScholarDigital Library
- Paolo Giacomazzi, Luigi Musumeci, Gabriella Saddemi, and Giacomo Verticale. Analytical methods for resource allocation and admission control with dual-leaky-bucket regulated traffic. In Proceedings of IEEE International Conference on Communications ICC, pages 499--505. IEEE, 2007.Google ScholarCross Ref
- Z. Gong, X. Gu, and J. Wilkes. Press: Predictive elastic resource scaling for cloud systems. In Proceedings of IEEE International Conference on Network and Service Management, 2010.Google Scholar
- Ori Hadary, Luke Marshall, Ishai Menache, Abhisek Pan, Esaias E. Greeff, David Dion, Star Dorminey, Shailesh Joshi, Yang Chen, Mark Russinovich, and Thomas Moscibroda. Protean: VM allocation service at scale. In 14th USENIX Symposium on Operating Systems Design and Implementation, OSDI, pages 845--861. USENIX Association, 2020.Google Scholar
- S. Islam, J. Keung, K. Lee, and A. Liu. Empirical prediction models for adaptive resource provisioning in the cloud. Future Generation Computer Systems, 2012.Google ScholarDigital Library
- Pawel Janus and Krzysztof Rzadca. SLO-aware colocation of data center tasks based on instantaneous processor requirements. In Proceedings of the 2017 Symposium on Cloud Computing (SoCC), pages 256--268. ACM, 2017.Google ScholarDigital Library
- Yuxuan Jiang, Mohammad Shahrad, David Wentzlaff, Danny H. K. Tsang, and Carlee Joe-Wong. Burstable instances for clouds: Performance modeling, equilibrium analysis, and revenue maximization. In 2019 IEEE Conference on Computer Communications (INFOCOM), pages 1576--1584. IEEE, 2019.Google ScholarDigital Library
- SeyedAli Jokar Jandaghi, Kaveh Mahdaviani, and Cristiana Amza. Virtual instance resource usage modeling: A method for efficient resource provisioning in the cloud. In Proceedings of IFIP/IEEE IM 2017 Workshop: 2nd International Workshop on Analytics for Network and Service Management (AnNet), pages 917--922, 2017.Google ScholarDigital Library
- A. Khan, X. Yan, S. Tao, and N. Anerousis. Workload characterization and prediction in the cloud: A multiple time series approach. In Proceedings of IEEE Network Operations and Management Symposium, 2012.Google ScholarCross Ref
- H. Nguyen, Z. Shen, X. Gu, S. Subbiah, and J. Wilkes. AGILE: Elastic distributed resource scaling for infrastructure-as-a-service. In Proceedings of International Conference on Autonomic Computing (ICAC), 2013.Google Scholar
- Hojin Park, Gregory R. Ganger, and George Amvrosiadis. More IOPS for less: Exploiting burstable storage in public clouds. In Amar Phanishayee and Ryan Stutsman, editors, 12th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud). USENIX Association, 2020.Google Scholar
- Riccardo Pinciroli, Ahsan Ali, Feng Yan, and Evgenia Smirni. CED-ULE+: resource management for burstable cloud instances using predictive analytics. IEEE Transactions on Network and Service Management, 18(1):945--957, 2021.Google ScholarDigital Library
- Olga Poppe, Tayo Amuneke, Dalitso Banda, Aritra De, Ari Green, Manon Knoertzer, Ehi Nosakhare, Karthik Rajendran, Deepak Shankargouda, Meina Wang, et al. Seagull: An infrastructure for load prediction and optimized resource allocation. arXiv preprint arXiv:2009.12922, 2020.Google Scholar
- N. Roy, A. Dubey, and A. Gokhale. Efficient autoscaling in the cloud using predictive models for workload forecasting. In Proceedings of IEEE International Conference on Cloud Computing, 2011.Google ScholarDigital Library
- Aakash Sharma, Saravanan Dhakshinamurthy, George Kesidis, and Chita R. Das. CASH: A credit aware scheduling for public cloud platforms. In Laurent Lefèvre, Stacy Patterson, Young Choon Lee, Haiying Shen, Shashikant Ilager, Mohammad Goudarzi, Adel Nadjaran Toosi, and Rajkumar Buyya, editors, 21st IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGrid), pages 227--236. IEEE, 2021.Google Scholar
- Z. Shen, S. Subbiah, X. Gu, and J. Wilkes. Cloudscale: Elastic resource scaling for multi-tenant cloud systems. In Proceedings of European Conference on Computer Systems (EuroSys), 2011.Google ScholarDigital Library
- Bo Sun, Yuxuan Jiang, and Danny H. K. Tsang. When burstable instances meet mobile computing: Performance modeling and economic analysis. In 40th IEEE International Conference on Distributed Computing Systems (ICDCS), pages 1179--1180. IEEE, 2020.Google ScholarCross Ref
- X. Sun, N. Ansari, and R.Wang. Optimizing resource utilization of a data center. IEEE Communications Surveys & Tutorials, 2016.Google ScholarDigital Library
- Luan Teylo, Luciana Arantes, Pierre Sens, and Lúcia Maria de A Drummond. Scheduling bag-of-tasks in clouds using spot and burstable virtual machines. IEEE Transactions on Cloud Computing, 11(1):984--996, 2021.Google ScholarCross Ref
- M. Tirmazi, A. Barker, N. Deng, M.E. Haque, Z.G. Qin, S. Hand, M. Harchol-Balter, and J. Wilkes. Borg: The next generation. In Proceedings of European Conference on Computer Systems (EuroSys), 2020.Google ScholarDigital Library
- A. Verma, L. Pedrosa, M. Korupolu, D. Oppenheimer, E. Tune, and J. Wilkes. Large-scale cluster management at google with borg. In Proceedings of European Conference on Computer Systems (EuroSys), 2015.Google ScholarDigital Library
- Cheng Wang, Bhuvan Urgaonkar, Aayush Gupta, George Kesidis, and Qianlin Liang. Exploiting spot and burstable instances for improving the cost-efficacy of in-memory caches on the public cloud. In Gustavo Alonso, Ricardo Bianchini, and Marko Vukolic, editors, Proceedings of the Twelfth European Conference on Computer Systems (EuroSys), pages 620--634. ACM, 2017.Google ScholarDigital Library
- Cheng Wang, Bhuvan Urgaonkar, Neda Nasiriani, and George Kesidis. Using burstable instances in the public cloud: Why, when and how? Proceedings of the ACM on Measurement and Analysis of Computing Systems, 1(1):11:1--11:28, 2017.Google ScholarDigital Library
Recommendations
Live Migration of Multiple Virtual Machines with Resource Reservation in Cloud Computing Environments
CLOUD '11: Proceedings of the 2011 IEEE 4th International Conference on Cloud ComputingVirtualization technology is currently becoming increasingly popular and valuable in cloud computing environments due to the benefits of server consolidation, live migration, and resource isolation. Live migration of virtual machines can be used to ...
Bayesian network-based Virtual Machines consolidation method
Efficient Virtual Machines (VMs) consolidation, as one of the primary methods for balancing between guaranteeing Quality of Service (QoS) and saving energy, is critical for data centers. Most existing VMs consolidation methods reallocate physical ...
Live gang migration of virtual machines
HPDC '11: Proceedings of the 20th international symposium on High performance distributed computingThis paper addresses the problem of simultaneously migrating a group of co-located and live virtual machines (VMs), i.e, VMs executing on the same physical machine. We refer to such a mass simultaneous migration of active VMs as "live gang migration". ...
Comments