The continuing growth of the cloud computing market has led to an unprecedented diversity of cloud services with different performance characteristics. To support service selection, researchers and practitioners conduct cloud performance benchmarking by measuring and objectively comparing the performance of different providers and configurations (e.g., instance types in different data center regions). In this tutorial, we demonstrate how to write performance tests for IaaS clouds using the Web-based benchmarking tool Cloud WorkBench (CWB). We will motivate and introduce benchmarking of IaaS cloud in general, demonstrate the execution of a simple benchmark in a public cloud environment, summarize the CWB tool architecture, and interactively develop and deploy a more advanced benchmark together with the participants.
Joel Scheuner (Chalmers | University of Gothenburg, Sweden) and Philipp Leitner (Chalmers | University of Gothenburg, Sweden)
This tutorials targets practitioners, researchers (academic and industrial), students, and developers that are interested in evaluating the performance of IaaS clouds. Tutorial participants are invited to follow easy steps in their web browser to execute a pre-configured benchmark in a public IaaS cloud. Interested tutorial participants will be given support to design, build, and execute a more advanced benchmark from scratch using a template. For the second part, some familiarity with the Bash shell and the Ruby programming language is beneficial but not required.
Personal laptop for participants who would like to follow the interactive part.
- IaaS cloud benchmarking (~20min): We present a general introduction into the area, with background and motivation.
- My first benchmark with CWB (~30min): We demonstrate how to modify an existing simple benchmark and execute it in a public IaaS cloud using the CWB web interface. Tutorial participants are invited to follow these steps in their web browser, and the presenters support the participants to do so.
- CWB Architecture Overview and selected previous results (~20min): We present a short overview of the basic architecture of CWB from a tool user perspective and summarize results of selected studies conducted with CWB.
- Building an advanced benchmark with CWB (~90min): In the second half of the tutorial, we introduce how to design and implement a more advanced benchmark from scratch, using Chef cookbooks to install dependencies and configure hooks to control the lifecycle of a benchmark. Interested tutorial participants will be given credentials and support to design, build, and execute such a benchmark from scratch using a template.