CHASE: Component High Availability-Aware Scheduler in Cloud Computing Environment

Cloud computing promises flexible integration of the compute capabilities for on-demand access through the concept of virtualization. However, uncertainties are raised regarding the high availability of the cloud-hosted applications. High availability is a crucial requirement for multi-tier applications providing business services for a broad range of enterprises. This paper proposes a novel component high availability-aware scheduling technique, CHASE, which maximizes the availability of applications without violating service level agreements with the end-users. Using CHASE, prior criticality analysis is conducted on applications to schedule them based on their impact on their execution environment and business functionality. This paper presents the advantages and shortcomings of CHASE compared to an optimal solution, Open Stack Nova scheduler, high availability-agnostic, and redundancy-agnostic schedulers. The evaluation results demonstrate that the proposed solution improves the availability of the scheduled components compared to the latter schedulers. CHASE prototype is also defined for runtime scheduling in Open Stack environment.

[1]  Pla Air.  ON SYSTEM RELIABILITY AND AVAILABILITY , 1987 .

[2]  Chen Jing,et al.  A dynamic and integrated load-balancing scheduling algorithm for Cloud datacenters , 2011, 2011 IEEE International Conference on Cloud Computing and Intelligence Systems.

[3]  Elliot K. Kolodner,et al.  Guaranteeing High Availability Goals for Virtual Machine Placement , 2011, 2011 31st International Conference on Distributed Computing Systems.

[4]  P. Parley.  What to Do, and How to Do It , 2011 .

[5]  Yiming Li,et al.  Software defined networking: State of the art and research challenges , 2014, Comput. Networks.

[6]  Sebastian Muller.  Openstack Operations Guide , 2016 .

[7]  Tom Fifield,et al.  OpenStack Operations Guide , 2014 .

[8]  Michael I. Jordan,et al.  Characterizing, modeling, and generating workload spikes for stateful services , 2010, SoCC '10.

[9]  Sherif Sakr,et al.  Incorporating Uncertainty into In-Cloud Application Deployment Decisions for Availability , 2013, 2013 IEEE Sixth International Conference on Cloud Computing.

[10]  Abdelkader H. Ouda,et al.  Resource allocation in a network-based cloud computing environment: design challenges , 2013, IEEE Communications Magazine.

[11]  Marcel Mongeau,et al.  Event-based MILP models for resource-constrained project scheduling problems , 2011, Comput. Oper. Res..

[12]  Fumio Machida,et al.  Redundant virtual machine placement for fault-tolerant consolidated server clusters , 2010, 2010 IEEE Network Operations and Management Symposium - NOMS 2010.

[13]  Jianhua Gu,et al.  A Scheduling Strategy on Load Balancing of Virtual Machine Resources in Cloud Computing Environment , 2010, 2010 3rd International Symposium on Parallel Architectures, Algorithms and Programming.

[14]  Calton Pu,et al.  Performance and availability aware regeneration for cloud based multitier applications , 2010, 2010 IEEE/IFIP International Conference on Dependable Systems & Networks (DSN).

[15]  Sherif Sakr,et al.  Improving Availability of Cloud-Based Applications through Deployment Choices , 2013, 2013 IEEE Sixth International Conference on Cloud Computing.

[16]  Xi Chen,et al.  An Availability-Aware Virtual Machine Placement Approach for Dynamic Scaling of Cloud Applications , 2012, 2012 9th International Conference on Ubiquitous Intelligence and Computing and 9th International Conference on Autonomic and Trusted Computing.

[17]  Abdallah Shami,et al.  High availability-aware optimization digest for applications deployment in cloud , 2015, 2015 IEEE International Conference on Communications (ICC).

[18]  Ravi Shankar,et al.  DynaPlan: Resource placement for application-level clustering , 2011, 2011 IEEE/IFIP 41st International Conference on Dependable Systems and Networks Workshops (DSN-W).

[19]  Abdallah Shami,et al.  NFV: state of the art, challenges, and implementation in next generation mobile networks (vEPC) , 2014, IEEE Network.