Research from Chair of Software Engineering featured on CACM Video Channel

Full automation of IT infrastructure and the delivery of efficient IT operations as billed services have been long-standing goals of the computing industry since at least the 1960s. A newcomer, serverless computing, emerged in the late 2010s with characteristics claimed to be different from those of established IT services. Even though serverless computing has gained significant attention in industry and academia over the past five years, there is still no consensus about its unique distinguishing characteristics and precise understanding of how these characteristics differ from classical cloud computing.

The article "Serverless Computing: What It Is, and What It Is Not?" seeks to dispel this confusion among others by proposing a refined definition capturing the essential conceptual characteristics of serverless computing as a paradigm, while putting the various terms around it into perspective. It starts by providing an analogy to intuitively illustrate serverless computing and how it compares to classical cloud computing. It then examines how the term serverless computing, and related terms, are used today while reviewing the historical evolution leading to serverless computing, starting with mainframe virtualization in the 1960s through to grid and cloud computing all the way up to today. Further, the article reviews existing cloud computing service models (including IaaS, PaaS, SaaS, CaaS, FaaS, and BaaS), discusses serverless computing applications and open challenges, and provides an outlook on the future of the serverless paradigm. Conceptual understanding of the state of the art coupled with clear and consistent terminology provide a basis for supporting interoperability between emerging cloud platforms as well as for future research driving the further advancement of the field.

The question of just what is serverless computing and how it differs from classical cloud computing was discussed extensively at a Dagstuhl Seminar we organized in 2021, bringing together around 50 experts from academia and industry, representing three communities of experts in computer systems, software engineering, and performance engineering. The discussions at the seminar sparked an effort to provide a new refined definition of serverless computing coupled with a long-term perspective on how the serverless paradigm fits in the space of existing and emerging cloud computing service models. The result has now been published togehter with a featured video by the Communication of the ACM (CACM):

Serverless Computing: What It Is, and What It Is Not? Kounev, Samuel; Herbst, Nikolas; Abad, Cristina L.; Iosup, Alexandru; Foster, Ian; Shenoy, Prashant; Rana, Omer; Chien, Andrew A.; in Commun. ACM (2023). 66(9) 80–92. [Open Access URL, Video]

Watch a video on the article: https://vimeo.com/849237573