QoSCloud

QoS-driven Cloud Computing and Storage Resource Management

System virtualization is an increasingly powerful technology that enables the emerging computing paradigms such as public and private cloud systems. It allows applications to be conveniently deployed along with their required execution environments through virtual machine (VMs), and supports them to flexibly share the underlying physical resources with strong isolation. However, there exists an increasingly urgent need for virtualized systems to deliver strong Quality of Service (QoS) guarantees to their hosted applications. Currently such systems can meet only coarse-grained and relaxed performance requirements, and their management considers only limited facets of an application’s multi-type resource usage. As a result, examples such as cloud systems cannot support QoS-based Service Level Agreements (SLA) with their hosted applications. The continued existence of the lack of strong QoS guarantees from virtualized systems presents a critical hurdle to their further adoption by applications and their support of more economical QoS-based SLAs. The objective of this project is to create a QoS-driven multi-type resource management system to support strong QoS guarantees for applications hosted on virtualized computing systems.

Resource management in virtualized systems remains a key challenge because of their intrinsically dynamic and complex nature, where the applications have dynamically changing workloads and virtual machines (VMs) compete for the shared resources in a convolved manner. To address this challenge, this project proposes a new resource management approach that can effectively capture the nonlinear behaviors in VM resource usages through fuzzy modeling and quickly adapt to the changes in the virtualized system through predictive control. The resulting fuzzy-model-predictive-control (FMPC) approach is capable of optimizing the VM-to-resource allocations according to high-level service differentiation or revenue maximization objectives.

Existing resource management solutions in datacenters and cloud systems typically treat VMs as black boxes when making resource allocation decisions. This project advocates the cooperation between VM host- and guest-layer schedulers for optimizing the resource management and application performance. It presents an approach to such cross-layer optimization upon fuzzy-modeling-based resource management. This approach exploits guest-layer application knowledge to capture workload characteristics and improve VM modeling, and enables the host-layer scheduler to feedback resource allocation decision and adapt guest-layer application configuration.

Participants

• Dr. Ming Zhao (faculty)
• Yitao Chen (PhD student)
• Sungho Hong (PhD student)
• Runyu Jin (PhD student)
• Wenji Li (PhD student)
• Michel Roger (PhD student; FIU B.S. and M.S., past Research Experiences for Undergraduates participant)
• Yiting Yao (PhD student)
• Qirui Yang (PhD student)
• Kaiqi Zhao (PhD student)

• Dulcardo Arteaga (PhD student – graduated; Now at Parallel Machines)
• Lixi Wang (PhD student – graduated; Now at Amazon)
• Yiqi Xu (PhD student – graduated, VMware Graduate Fellow; Now at VMware)
• Ragini Sharma (MS student – graduated; Now at Paypal)
• Andrew Nguyen (Research Experiences for Undergraduates participant; Now at Northrop Grumman)
• Kyler Butler (B.S. Student, Research Experiences for Undergraduates participant)
• Eduardo Castillo (B.S. Student – graduated, Research Experiences for Undergraduates participant)
• Francois D’Ugard (B.S. student – graduated, Research Experiences for Veterans participant; Now at IBM)
• Terry Henderson (B.S. student – graduated, Research Experiences for Veterans participant)
• Steven Igneti (B.S. Student – graduated, Research Experiences for Undergraduates participant)
• Gregory Jean-Baptise (McKnight Doctoral Fellow, FIU B.S., past Research Experiences for Undergraduates participant; Now at VMware)
• Bryan Jimenez (B.S. student – graduated, Research Experiences for Veterans participant; Now at of University of Miami)
• Peter Reidy (B.S. Student – graduated, Research Experiences for Undergraduates participant)
• Olena Tkachenko (B.S. Student – graduated, Research Experiences for Undergraduates participant)
• Kyle Zinke (B.S. Student, Research Experiences for Undergraduates participant)

Publications

• Q. Yang, R. Jin, and M. Zhao, “SmartDedup: Optimizing Deduplication for Resource-constrained Devices,” Proceedings of the USENIX Annual Technical Conference (USENIX ATC’19), July 2019.
• Y. Chen, K. Zhao, B. Li, and M. Zhao, “Exploring the Use of Synthetic Gradients for Distributed Deep Learning across Cloud and Edge Resources,” Proceedings of the USENIX Workshop on Hot Topics in Edge Computing (HotEdge), July 2019.
• J. Fu, Y. Lu, J. Shu, G. Liu, and M. Zhao, “CowCache: Effective Flash Caching for Copy-on-Write Virtual Disks,” Cluster Computing, June 2019.
• Biookaghazadeh, Y. Chen, K. Zhao, and M. Zhao, “KnowledgeNet: Disaggregated and Distributed Training and Serving of Deep Neural Networks,” Proceedings of the USENIX Conference on Operational Machine Learning (OpML’19), May 2019.
• M. Zhao and Y. Xu, “vPFS+: Managing I/O Performance for Diverse HPC Applications,” Proceedings of the 35th International Conference on Massive Storage Systems and Technology (MSST 2019), May 2019.
• P. Zuo, Y. Hua, M. Zhao, W. Zhou, and Y. Guo, “Write Deduplication and Hash Mode Encryption for Secure Nonvolatile Main Memory,” IEEE Micro, Volume 39, Pages 44-51, January/February 2019.
• P. Zuo, Y. Hua, M. Zhao, W. Zhou and Y. Guo, “Improving the Performance and Endurance of Encrypted Non-volatile Main Memory through Deduplicating Writes,” Proceedings of the 51st IEEE/ACM International Symposium on Microarchitecture (MICRO), October 2018.
• S. Biookaghazadeh, M. Zhao, and F. Ren, “Are FPGAs Suitable for Edge Computing?” Proceedings of the USENIX Workshop on Hot Topics in Edge Computing (HotEdge), July 2018.
• G. Vietri, L. V. Rodriguez, W. A. Martinez, S. Lyons, J. Liu, R. Rangaswami, M. Zhao, G. Narasimhan, “Driving Cache Replacement with ML-based LeCaR,” Proceedings of the 10th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage), July 2018.
• R. Sharma, S. Biookaghazadeh, and M. Zhao. “Are Existing Knowledge Transfer Techniques Effective For Deep Learning on Edge Devices?” Proceedings of the IEEE International Conference on Edge Computing (EDGE), July 2018.
• M. Zhao and J. Cabrera, “RTVirt: Enabling Time-sensitive Computing on Virtualized Systems through Cross-layer CPU Scheduling,” Proceedings of the European Conference on Computer Systems (EuroSys), April 2018.
• M. Zhao, L. Wang, Y. Lv, and J. Xu, “Cross-layer Optimization for Virtual Machine Resource Management,” Proceedings of the IEEE International Conference on Cloud Engineering (IC2E), April 2018.
• V. Tarasov, L. Rupprecht, D. Skourtis, A. Warke, D. Hildebrand, M. Mohamed, N. Mandagere, W. Li, R. Rangaswami, and M. Zhao, “In Search of the Ideal Storage Configuration for Docker Containers,” Proceedings of the 1st Workshop on Autonomic Management of Large Scale Container-based Systems (AMLCS), September 2017.
• S. Biookaghazadeh, S. Zhou, and M. Zhao, “Kaleido: Enabling Efficient Scientific Data Processing on Big-Data Systems,” Proceedings of the 12th International Conference on Networking, Architecture, and Storage (NAS), August 2017.
• Y. Xu and M. Zhao, “IBIS: Interposed Big-data I/O Scheduler,” Proceedings of the 25th International Symposium on High-Performance Parallel and Distributed Computing, May 2016.
• S. Biookaghazadeh, Y. Xu, S. Zhou, and M. Zhao, “Enabling Scientific Data Storage and Processing on Big-data Systems,” Proceedings of the Big Data in the Geosciences Workshop (co-held with 2015 IEEE International Big Data Conference), October 2015.
• W. Li, G. Jean-Baptise, J. Riveros, G. Narasimhan, T. Zhang, and M. Zhao, “CacheDedup: In-line Deduplication for Flash Caching,” Proceedings of the 14th USENIX Conference on File and Storage Technologies (FAST’16), February 2016.
• D. Arteaga, J. Cabrera, J. Xu, S. Sundararaman, and M. Zhao, “CloudCache: On-demand Flash Cache Management for Cloud Computing,” Proceedings of the 14th USENIX Conference on File and Storage Technologies (FAST’16), February 2016.
• L. Wang, J. Xu, and M. Zhao, “QoS-driven Cloud Resource Management through Fuzzy Model Predictive Control,” Proceedings of the 12th International Conference on Autonomic Computing (ICAC), July 2015.
• S. Kundu, R. Rangaswami, M. Zhao, A. Gulati, and K. Dutta, “Revenue Driven Resource Allocation for Virtualized Data Centers,” Proceedings of 12th International Conference on Autonomic Computing (ICAC), July 2015.
• M. Zhao, F. D’Ugard, K. Kwiat, and C. Kamhoua, “Multi-level VM Replication based Survivability for Mission-critical Cloud Computing,” Proceedings of the 1st IEEE/IFIP Workshop on Security for Emerging Distributed Network Technologies (co-located with IEEE/IFIP International Symposium on Integrated Network Management), May 2015.
• J. Li, K. Zhao, X. Zhang, J. Ma, M. Zhao, and T. Zhang, “How Much Can Data Compressibility Help to Improve NAND Flash Memory Lifetime?” (to appear) Proceedings of the 13th USENIX Conference on File and Storage Technologies (FAST’15), February 2015.
• M. Roger, Y. Xu, and M. Zhao, “BigCache for Big-data Systems,” Proceedings of the IEEE International Conference on Big Data (BigData2014), October 2014.
• D. Otstott, N. Evans, L. Ionkov, M. Zhao, and M. Lang, “Enabling Composite Applications through an Asynchronous Shared Memory Interface,” Proceedings of the IEEE International Conference on Big Data (BigData2014), October 2014.
• Y. Lu, M. Zhao, G. Zhao, L. Wang, and N. Rishe, “v-TerraFly: Large Scale Distributed Spatial Data Visualization with Autonomic Resource Management,” Journal of Big Data, 1:4, Pages: 1-19, 2014.
• Dulcardo Arteaga and Ming Zhao, “Client-side Flash Caching for Cloud Systems,” 7th ACM International Systems and Storage Conference, June 2014.
• Douglas Otstott, Jorge Cabrera, and Ming Zhao, “A Host-side Integrated Flash Scheduler for Solid State Drives,” 12th USENIX Conference on File and Storage Technologies, February 2014. (Work in Progress)
• Gregory Jean-Baptiste, Dulcardo Arteaga, and Ming Zhao, “Inline Deduplication for Storage Caching,” 12th USENIX Conference on File and Storage Technologies, February 2014. (Work in Progress)
• Yun Lu, Ming Zhao, Guangqiang Zhao, Lixi Wang, Naphtali Rishe, “v-TerraFly: Large Scale Distributed Spatial Data Visualization with Autonomic Resource Management,” Journal Of Big Data, January 2014.
• Yiqi Xu, Adrian Suarez, and Ming Zhao, “IBIS: Interposed Big-data I/O Scheduler,” 22nd ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC2013), June 2013. (Short Paper)
• Dulcardo Arteaga and Ming Zhao, “Trace Analysis for Block-level Caching in Cloud Computing Systems,” 11th USENIX Conference on File and Storage Technologies (FAST2013), February 2013 (Work in Progress)
• Ricardo Koller, Leonardo Marmol, Raju Rangaswami, Swaminathan Sundararaman, Nisha Talagala, and Ming Zhao, “Write Policies for Host-side Flash Caches,” 11th USENIX Conference on File and Storage Technologies (FAST’13), February 2013
• Lixi Wang, Jing Xu, and Ming Zhao, “Modeling VM Performance Interference with Fuzzy MIMO Model,” 7th International Workshop on Feedback Computing (FeedbackComputing, co-held with ICAC2012), September 2012.
• Lixi Wang, Jing Xu, and Ming Zhao, “Application-aware Cross-layer Virtual Machine Resource Management,” 9th International Conference on Autonomic Computing (ICAC), September 2012.
• Dulcardo Arteaga, Douglas Otstott, and Ming Zhao, “Dynamic Block-level Cache Management for Cloud Computing Systems,” 10th USENIX Conference on File and Storage Technologies (FAST’12), March 2012. (Work in Progress)
• Sajib Kundu, Raju Rangaswami, Ajay Gulati, Ming Zhao, Kaushik Dutta, “Modeling Virtualized Applications using Machine Learning Techniques,” 8th Annual International Conference on Virtual Execution Environments (VEE 2012), March 2012.
• Lixi Wang, Jing Xu, Ming Zhao, Yicheng Tu, Jose Fortes, “Fuzzy Modeling based Resource Management for Virtualized Database Systems,” 19th Annual Meeting of the IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS 2011), July 2011.
• Yangyang Wu, Ming Zhao, “Performance Modeling of Virtual Machine Live Migration,” 4th IEEE International Conference on Cloud Computing (CLOUD 2011), July 2011.
• Lixi Wang, Jing Xu, Ming Zhao, Jose Fortes, “Adaptive Virtual Resource Management with Fuzzy Model Predictive Control,” 6th International Workshop on Feedback Control Implementation and Design in Computing Systems and Networks (FeBID, co-held with ICAC’11), June 2011.
• Lixi Wang, Jing Xu, Ming Zhao, Jose Fortes, “Adaptive Virtual Resource Management with Fuzzy Model Predictive Control,” 8th International Conference on Autonomic Computing (ICAC’11), June 2011. (Short Paper)
• Dulcardo Arteaga, Ming Zhao, Chen Liu, Pollawat Thanarungroj, Lichen Weng, “Cooperative Virtual Machine Scheduling on Multi-core Multi-threading Systems — A Feasibility Study,” Workshop on Micro Architectural Support for Virtualization, Data Center Computing, and Cloud (MASVDC, co-held with MICRO 2010), December 2010.
• Sajib Kundu, Raju Rangaswami, Kaushik Dutta, and Ming Zhao, “Application Performance Modeling in a Virtualized Environment,” 16th IEEE International Symposium on High-Performance Computer Architecture (HPCA-16), January 2010.
• Jing Xu, Ming Zhao, José A. B. Fortes, Robert Carpenter, and Mazin Yousif, “Cooperative Autonomic Management in Dynamic Distributed Systems,” 11th International Symposium on Stabilization, Safety, and Security of Distributed Systems (SSS 2009), November 2009.
• Lixi Wang, Jing Xu, Ming Zhao, Yicheng Tu, Jose Fortes, “Autonomic Resource Management for Virtualized Database Hosting Systems,” Technical Report 2009-07-01, School of Computing and Information Sciences, Florida International University, July 2009.
• Jing Xu, Ming Zhao, and José A. B. Fortes, “Autonomic Resource Management in Virtualized Data Centers Using Fuzzy-logic-based Control,” Cluster Computing, Vol. 11, No. 3, Pages: 213-227, September 2008.

 

Education Activities

• CloudHackathon
• vMoodle – Cloud-based Online Learning System
• VISA session at FIU Engineering Expo 2014

 

Acknowledgement

This material is based upon work supported by the National Science Foundation CAREER award CNS-1253944/CNS-1619653.