The Great Puzzle: Cloud Functions, Data, Services, Less Servers and More Insights
ABSTRACT: Many Function-as-a-Service application models have been proposed and implemented in recent years to perform scalable event-driven data processing. Ranging from pure cloud functions to more distributed approaches involving continuums and osmotic infrastructures, they areparticularly well-suited for dynamic environments with varying loads. From an applied sciences perspective, arriving at an application in production from a hypothetical application model and code base remains the largest challenge. Based on recent industry collaborations, the talk explains pieces of the puzzle to achieve that goal. The anticipated advances include the need to know more about the computing environment and to formalise this knowledge, to provide better service-oriented access to data representing these insights for greater system autonomy, and to interface better with leading cloud-native platforms. In the long term, we need to introduce more liquidity for adaptive improvements in microservice composition and placement so that new knowledge can be accurately reflected in an optimal system configuration.
Josef Spillner is head of the Service Prototyping Lab and associate professor at Zurich University of Applied Sciences in Switzerland. With his team, he advances computing paradigms around the topics of cloud-native and post-cloud applications, serverless computing, messaging and microbilling. Moreover, he explores novel ways to publish practically useful research output. Earlier in his career, he published a doctoral dissertation about metaquality of services and a habilitation treatise about stealth computing in multi-cloud environments.