Wednesday, 13 December 2017

Session 10: Performance in Cloud, Virtualized and Multi-Core Systems

Towards Building Performance Models for Data-intensive Workloads in Public Clouds

Authors:

Rizwan Mian (Queen's University)
Patrick Martin (Queen's University)
Farhana Zulkernine (Queen's University)
Jose Luis Vazquez-Poletti (Universidad Complutense de Madrid)

Abstract:

The cloud computing paradigm provides the “illusion” of infinite resources and, therefore, becomes a promising candidate for large-scale data-intensive computing. In this paper, we explore experiment-driven performance models for data-intensive workloads executing in an infrastructure-as-a-service (IaaS) public cloud. The performance models help in predicting the workload behaviour, and serve as a key component of a larger framework for resource provisioning in the cloud. We determine a suitable prediction technique after comparing popular regression methods. We also enumerate the variables that impact variance in the workload performance in a public cloud. Finally, we build a performance model for a multi-tenant data service in the Amazon cloud. We find that a linear classifier is sufficient in most cases. On a few occasions, a linear classifier is unsuitable and non-linear modeling is required, which is time consuming. Consequently, we recommend that a linear classifier be used in training the performance model in the first instance. If the resulting model is unsatisfactory, then non-linear modeling can be carried out in the next step.

DOI: 10.1145/2479871.2479908

Full text: PDF

[#][]

vPerfGuard: An Automated Model-Driven Framework for Application Performance Diagnosis in Consolidated Cloud Environments

Authors:

Pengcheng Xiong (Georgia Institute of Technology)
Calton Pu (Georgia Institute of Technology)
Xiaoyun Zhu (VMware Inc.)
Rean Griffith (VMware Inc.)

Abstract:

Many business customers hesitate to move all their applications to the cloud due to performance concerns. White-box diagnosis relies on human expert experience or performance troubleshooting “cookbooks” to find potential performance bottlenecks. Despite wide adoption, the scalability and adaptivity of such approaches remain severely constrained, especially in a highly-dynamic, consolidated cloud environment. Leveraging the rich telemetry collected from applications and systems in the cloud, and the power of statistical learning, vPerfGuard complements the existing approaches with a model-driven framework by: (1) automatically identifying system metrics that are most predictive of application performance, and (2) adaptively detecting changes in the performance and potential shifts in the predictive metrics that may accompany such a change. Although correlation does not imply causation, the predictive system metrics point to potential causes that can guide a cloud service provider to zero in on the root cause.

We have implemented vPerfGuard as a combination of three modules: a sensor module, a model building module, and a model updating module. We evaluate its effectiveness using different benchmarks and different workload types, specifically focusing on various resource (CPU, memory, disk I/O) contention scenarios that are caused by workload surges or“noisy neighbors”. The results show that vPerfGuard automatically points to the correct performance bottleneck in each scenario, including the type of the contended resource and the host where the contention occurred.

DOI: 10.1145/2479871.2479909

Full text: PDF

[#][]

Predictive Performance Modeling of Virtualized Storage Systems using Optimized Statistical Regression Techniques

Authors:

Qais Noorshams (Karlsruhe Institute of Technology)
Dominik Bruhn (Karlsruhe Institute of Technology)
Samuel Kounev (Karlsruhe Institute of Technology)
Ralf Reussner (Karlsruhe Institute of Technology)

Abstract:

Modern virtualized environments are key for reducing the operating costs of data centers. By enabling the sharing of physical resources, virtualization promises increased resource efficiency with decreased administration costs. With the increasing popularity of I/O-intensive applications, however, the virtualized storage used in such environments can quickly become a bottleneck and lead to performance and scalability issues. Performance modeling and evaluation techniques applied prior to system deployment help to avoid such issues. In current practice, however, virtualized storage and its performance-influencing factors are often neglected or treated as a black-box. In this paper, we present a measurement-based performance prediction approach for virtualized storage systems based on optimized statistical regression techniques. We first propose a general heuristic search algorithm to optimize the parameters of regression techniques. Then, we apply our optimization approach and create performance models using four regression techniques. Finally, we present an in-depth evaluation of our approach in a real-world representative environment based on IBM System z and IBM DS8700 server hardware. Using our optimized techniques, we effectively create performance models with less than 7% prediction error in the most typical scenario. Furthermore, our optimization approach reduces the prediction error by up to 74%.

DOI: 10.1145/2479871.2479910

Full text: PDF

[#][]

Experimental Analysis of Task-Based Energy Consumption in Cloud Computing Systems

Authors:

Feifei Chen (Swinburne University of Technology)
John Grundy (Swinburne University of Technology)
Yun Yang (Swinburne University of Technology)
Jean-Guy Schneider (Swinburne University of Technology)
Qiang He (Swinburne University of Technology)

Abstract:

Cloud computing delivers IT solutions as a utility to users. One consequence of this model is that large cloud data centres consume large amounts of energy and produce significant carbon footprints. A common objective of cloud providers is to develop resource provisioning and management solutions that minimise energy consumption while guaranteeing Service Level Agreements (SLAs). In order to achieve this objective, a thorough understanding of energy consumption patterns in complex cloud systems is imperative. We have developed an energy consumption model for cloud computing systems. To operationalise this model, we have conducted extensive experiments to profile the energy consumption in cloud computing systems based on three types of tasks: computation-intensive, data-intensive and communication-intensive tasks. We collected fine-grained energy consumption and performance data with varying system configurations and workloads. Our experimental results show the correlation coefficients of energy consumption, system configuration and workload, as well as system performance in cloud systems. These results can be used for designing energy consumption monitors, and static or dynamic system-level energy consumption optimisation strategies for green cloud computing systems.

DOI: 10.1145/2479871.2479911

Full text: PDF

[#][]