Serverless computing has become increasingly popular since Amazon released AWS Lambda in 2014. Its appeal lies in the ability to write arbitrary code and deploy it at scale with the “click of a button”, in some sense realizing the original promise of cloud computing. While existing serverless infrastructure works well for a certain class of applications, it also has a number of significant limitations that cripple its generality. For example, stateful applications are forced into well-known anti-patterns like data shipping. In this talk, we briefly discuss both the benefits and shortcomings of existing serverless offerings. We then project forward to the future and highlight challenges that must be overcome to realize truly general-purpose serverless computing, as well as our efforts to get there.
Chenggang Wu is a fourth-year Ph.D. student at UC Berkeley working with Professor Joseph M. Hellerstein. His research interests lie in data-centric systems and distributed systems. Prior to joining Berkeley, he obtained his B.S. degree in computer science from Brown University in 2015.