Performance Modeling to Support Multi-tier Application Deployment to Infrastructure-as-a-Service Clouds

Lloyd, Wes; Pallickara, Shrideep; David, Olaf; Lyon, Jim; Arabi, Mazdak; Rojas, Ken

doi:10.1109/ucc.2012.20

Cited by 19 publications

(12 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lloyd et al [8] used physical and virtual machine resource utilization statistics to build performance models. They evaluated eighteen individual resource utilization statistics by predicting service execution time using several modeling approaches.…”

Section: Virtualized I/omentioning

confidence: 99%

Network Aware Scheduling for Virtual Machine Workloads with Interference Models

Verboven

Vanmechelen

Broeckhove

2015

IEEE Trans. Serv. Comput.

View full text Add to dashboard Cite

Modern data centers use virtualization as a means to increase utilization of increasingly powerful multi-core servers. Applications often require only a fraction of the resources provided by modern hardware. Multiple concurrent workloads are therefore required to achieve adequate utilization levels. Current virtualization solutions allow hardware to be partitioned into Virtual Machines with appropriate isolation on most levels. However, unmanaged consolidation of resource intensive workloads can still lead to unexpected performance variance. Measures are required to avoid or reduce performance interference and provide predictable service levels for all applications. In this paper, we identify and reduce network-related interference effects using performance models based on the runtime characteristics of virtualized workloads. We increase the applicability of existing training data by adding network-related performance metrics and benchmarks. Using the extended set of training data, we predict performance degradation with existing modeling techniques as well as combinations thereof. Application clustering is used to identify several new network-related application types with clearly defined performance profiles. Finally, we validate the added value of the improved models by introducing new scheduling techniques and comparing them to previous efforts. We demonstrate how the inclusion of network-related parameters in performance models can significantly increase the performance of consolidated workloads.

show abstract

Section: Virtualized I/omentioning

confidence: 99%

Network Aware Scheduling for Virtual Machine Workloads with Interference Models

Verboven

Vanmechelen

Broeckhove

2015

IEEE Trans. Serv. Comput.

View full text Add to dashboard Cite

show abstract

“…The objective of the ILP is to find the minimum number and locations of the data centers that satisfy the above constraint [8]. 1 In reality, data center locations are determined by a number of additional practical factors, e.g., land prices, electricity charges etc.…”

Section: A Data Center Locationsmentioning

confidence: 99%

“…Let h (1) ij and h (2) ij denote the hop count of the shortest path and the second shortest path between node i and node j respectively. The range between node i and node j is defined as:…”

Section: A Data Center Locationsmentioning

confidence: 99%

“…As both data center and network technologies become more mature, the next generation cloud applications are demanding improved end-to-end performance assurance from the cloud infrastructure while maintaining a competitive cost of service [1]. As a result, end-to-end performance assurance in terms of application response time, service availability, achievable throughput, etc., becomes a key performance indicator in the cloud system.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Planning and online resource allocation for the multi-resource cloud infrastructure

Wang

Razo

Tacca

et al. 2014

2014 IEEE International Conference on Communications (ICC)

View full text Add to dashboard Cite

Given the geographic complexity of the cloud infrastructure distribution, both offline data center dimensioning and link dimensioning as well as online resource allocation strategies play a significant role in enhancing end-to-end performance.In this paper, a three-step offline capacity assignment approach is proposed to select the data center locations, dimension the data centers and network links in an arbitrary mesh topology. Then the performance of the capacity assignment approach is investigated using a number of joint allocation strategies.Simulation results show that by using joint online resource allocation strategies, up to 50% higher load can be offered to the system at a given blocking probability. In addition, the combination of using both optimized capacity assignment approach and joint allocation strategy can reduce the link bandwidth overprovisioning. Results also show that the joint allocation strategy is more robust against sub-optimal offline capacity assignment.

show abstract

“…In [6], authors propose two load balancing algorithms for performance improvement, while an interesting analysis of the relation between performance and different types of workload is presented in [7]. However, neither energy efficiency nor dependability are considered.…”

Section: Related Workmentioning

confidence: 99%

Cost-Benefit Analysis of Virtualizing Batch Systems: Performance-Energy-Dependability Trade-Offs

Cinque

Cotroneo

Frattini

et al. 2013

2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing

View full text Add to dashboard Cite

Performance, energy efficiency, and dependability are key characteristics of batch systems, which can be differently affected when adopting virtualization. Scientific literature usually analyzes the variation with respect to different configurations of one characteristic, or the trade-off between two. In this paper, instead, we assess the impact of virtualization encompassing all of them. Results show that the joint analysis helps in finding the proper tuning of the system for balancing costs and benefits due to virtualization and related techniques.

show abstract

Performance Modeling to Support Multi-tier Application Deployment to Infrastructure-as-a-Service Clouds

Cited by 19 publications

References 14 publications

Network Aware Scheduling for Virtual Machine Workloads with Interference Models

Network Aware Scheduling for Virtual Machine Workloads with Interference Models

Planning and online resource allocation for the multi-resource cloud infrastructure

Cost-Benefit Analysis of Virtualizing Batch Systems: Performance-Energy-Dependability Trade-Offs

Contact Info

Product

Resources

About