Tutorial 1: Mo April 19 – 14h-17h CEST
Tutorial Title: Performance Modeling and Engineering for Microservices and Serverless Applications with the RADON Framework
Tutorial Abstract
This tutorial presents the performance engineering approach for microservices and serverless applications being developed as part of the RADON project (http://radon-h2020.eu). First, we will introduce microservices and serverless FaaS applications, and performance engineering challenges in this context. The focus will then shift to how these applications can be modeled using TOSCA and how different performance specifications can be integrated into such models. After that, the tutorial will discuss how such models are utilized to determine optimal decomposition strategies of serverless functions. Once the application gets deployed, following the DevOps practice, it is necessary to generate and maintain performance test cases for continuous integration and deployment. Finally, the tutorial focuses on giving theoretical baselines of resource management problems that arise in this area.
Presenter Contact information
A. Alnafessah 1, G. Casale 1, T. F. Düllmann 2, A. U. Gias 1, A. van Hoorn 2, M. Wurster 2, L. Zhu 1
- 1. Imperial College London, UK
- 2. University of Stuttgart, Germany
Link:
Tutorial 2: Mo April 19 – 18h-21h CEST
Tutorial Title: Enhancing Observability of Serverless Computing with the Serverless Application Analytics Framework
Tutorial Abstract
This tutorial will introduce serverless software development with Function-as-a-Service (FaaS) platforms. These platforms automate many features of distributed software that are notoriously difficult to implement including high availability, fault tolerance, and automatic scaling while only billing for the actual runtime of functions in sub-second intervals. Additionally these platforms support agile software development by helping programmers enhance productivity by supporting rapid development and deployment of web services and microservices. We will provide an overview and feature comparison to describe the state-of-the-art of existing Function-as-a-Service serverless platforms including cloud based and open-source platforms and discuss current challenges and limitations. We will then introduce the Serverless Application Analytics Framework (SAAF), an analytics framework that supports profiling workload performance, resource utilization, and infrastructure to enable performance and cost characterizations of serverless software deployments. SAAF, and its set of supporting tools, has been designed to improve the observability of serverless software deployments to enable developers to better evaluate design trade-offs. SAAF provides tools to profile software deployments to AWS Lambda, Google Cloud Functions, IBM Cloud Functions, and Azure Functions written in Java, Python, Node.js/Javascript, and BASH.
Presenter Contact information
- Robert Cordingly rcording@uw.edu School of Engineering and Technology University of Washington Tacoma
- Wes J. Lloyd wlloyd@uw.edu School of Engineering and Technology University of Washington Tacoma
Robert Cordingly - Brief Biography
Robert Cordingly is a PhD student and research assistant at the University of Washington Tacoma in the Computer Science and Systems program. His research focuses on serverless cloud computing, cloud performance analysis and cost modeling, and serverless application design. He is the primary developer for SAAF, FaaS Runner, and other supporting tools. With his tools and research, he aims to aid developers and researchers in making educated choices to improve application performance and cost on cloud computing platforms.
Wes Lloyd - Brief Biography
Wes Lloyd is an assistant professor in the Computer Science and Systems program at the School of Engineering and Technology at the University of Washington Tacoma. Wes teaches classes in cloud computing and systems. His research interests span cloud computing, distributed systems, and software engineering. His research aims to help scientists and practitioners better leverage cloud computing technologies to improve performance, availability, and observability of software deployments to maximize efficiency and lower hosting costs.
Link:
