Lambada: Interactive Data Analytics on Cold Data Using Serverless Cloud Infrastructure
Lambada is presented, a serverless distributed data processing framework designed to explore how to perform data analytics on serverless computing, and which scenarios serverless makes sense from an economic and performance perspective.
Abstract
The promise of ultimate elasticity and operational simplicity of serverless\ncomputing has recently lead to an explosion of research in this area. In the\ncontext of data analytics, the concept sounds appealing, but due to the\nlimitations of current offerings, there is no consensus yet on whether or not\nthis approach is technically and economically viable. In this paper, we\nidentify interactive data analytics on cold data as a use case where serverless\ncomputing excels. We design and implement Lambada, a system following a purely\nserverless architecture, in order to illustrate when and how serverless\ncomputing should be employed for data analytics. We propose several system\ncomponents that overcome the previously known limitations inherent in the\nserverless paradigm as well as additional ones we identify in this work. We can\nshow that, thanks to careful design, a serverless query processing system can\nbe at the same time one order of magnitude faster and two orders of magnitude\ncheaper compared to commercial Query-as-a-Service systems, the only alternative\nwith similar operational simplicity.\n