Tuesday, 12 December 2017

Session 7: Virtualized Environments

IO Performance Prediction in Consolidated Virtualized Environments

Authors:

Stephan Kraft (SAP Research)
Giuliano Casale (Imperial College London)
Diwakar Krishnamurthy (University of Calgary)
Des Greer (Queen's University Belfast)
Peter Kilpatrick (Queen's University Belfast)

Abstract:

We propose a trace-driven approach to predict the performance degradation of disk request response times due to storage device contention in consolidated virtualized environments. Our performance model evaluates a queueing network with fair share scheduling using trace-driven simulation. The model parameters can be deduced from measurements obtained inside Virtual Machines (VMs) from a system where a single VM accesses a remote storage server. The parameterized model can then be used to predict the effect of storage contention when multiple VMs are consolidated on the same virtualized server. The model parameter estimation relies on a search technique that tries to estimate the splitting and merging of blocks at the the Virtual Machine Monitor (VMM) level in the case of multiple competing VMs. Simulation experiments based on traces of the Postmark and FFSB disk benchmarks show that our model is able to accurately predict the impact of workload consolidation on VM disk IO response times.

DOI: 10.1145/1958746.1958789

Full text: PDF

[#][]

Virt-LM: A Benchmark for Live Migration of Virtual Machine

Authors:

Dawei Huang (Zhejiang University)
Deshi Ye (Zhejiang University)
Qinming He (Zhejiang University)
Jianhai Chen (Zhejiang University)
Kejiang Ye (Zhejiang University)

Abstract:

Virtualization technology has been widely applied in data centers and IT infrastructures, with advantages of server consolidation and live migration. Through live migration, data centers could flexibly move virtual machines among different physical machines to balance workloads, reduce energy consumption and enhance service availability.

Today's data centers can grow to a huge scale. This im- plies that frequent live migration would be desireble for the economic use of hardware resources. Then, the performance of the live migration strategy will be an issue. So, we need a reliant evaluation method to choose the software and hardware environments that will produce the best live migration performance.

However, there is not a complete live migration bench- mark available currently. In addition, the existing evaluation methodologies select different metrics, different workloads and different test means. Thus, it is difficult to compare their results.

In this paper we first survey the current research and their evaluation methods on live migration. We then summarize the critical issues for the live migration evaluation and also raise other unreported potential problems.

We propose our solutions and present an implementation in our live migration benchmark -- Virt-LM. This is a benchmark for comparing live migration performance among different software and hardware environments in a data center scenario. We detail its design and provide some experimental results to validate its effectiveness.

 

DOI: 10.1145/1958746.1958790

Full text: PDF

[#][]

Dynamic VM Migration: Assessing Its Risks & Rewards Using a Benchmark

Authors:

Krishnamurthy Srinivasan (Intel Corporation)
Sterlan Yuuw (Intel Corporation)
Tom J. Adelmeyer (Intel Corporation)

Abstract:

Dynamic migration of virtual machines (VMs) across physical servers has the potential to increase the utilization of the servers and hence drive down the data center costs. However, IT practitioners are leery of using this capability for increasing resource utilization due to concerns about the impact of such migrations on the performance of the applications, particularly the response times seen by the users of the applications. The relative newness to the industry of many of the tools used to automate VM migrations for resource utilization; data from researchers; as well as the recommendations from some analysts justify such caution and warrant quantifying the risks as well as potential rewards before deciding how aggressively this capability should be adopted. This paper will discuss the requirements for a benchmark to be used for such quantification. We will also discuss adaptations to SPECvirt_sc2010*, originally developed as a single server benchmark, to meet these requirements. We will also present risk-reward quantifications obtained using this benchmark for a simple case and the broader use of the benchmark for other cases.

DOI: 10.1145/1958746.1958791

Full text: PDF

[#][]

Performance Evaluation for Software Migration

Authors:

Issam Al-Azzoni (INRIA)
Lei Zhang (McMaster University)
Douglas G. Down (McMaster University)

Abstract:

Advances in technology and economical pressure have forced many organizations to consider the migration of their legacy systems to newer platforms. Legacy systems typically provide mission critical services vital for an organization's business needs. These systems are usually very large and highly complex with little or no documentation. Furthermore, fewer people can understand and maintain these systems. While several techniques exist to verify the functionality of the migrated system, the literature is still lacking methods to effectively assess the performance impact of software migration. In this paper, we propose a new method designed specifically to address performance evaluation in software migration projects. The new method uses simple models and incorporates techniques for model validation and resource demand mapping for performance evaluation and capacity planning.

DOI: 10.1145/1958746.1958792

Full text: PDF

[#][]

Modular Performance Modelling for Mobile Applications

Authors:

Niaz Arijo (University of Leicester)
Reiko Hecke (University of Leicester)
Mirco Tribastone (Ludwig-Maximilians-Universität)
Stephen Gilmore (University of Edinburgh)

Abstract:

We propose a model-based approach to analysing the performance of mobile applications where physical mobility and state changes are modelled by graph transformations from which a model in the Performance Evaluation Process Algebra (PEPA) is derived. To fight scalability problems with state space generation we adopt a modular solution where the graph transformation system is decomposed into views, for which labelled transition systems (LTS) are generated separately and later synchronised in PEPA. We demonstrate that the result of this modular analysis is equivalent to that of the monolithic approach and evaluate practicality and scalability by means of a case study.

DOI: 10.1145/1958746.1958793

Full text: PDF

[#][]