Network Resource Isolation in Serverless Cloud Function Service

Abstract

Serverless computing and function execution using cloud computing services are currently the subject considerable attention from both academia and industry. One of the reasons for the success of serverless computing is its straightforward interface that allows users to control the size of the memory allocated for the run-time of a function. However, this approach may result in the abstraction of too much information, and users cannot predict how their applications will perform, especially for the network resource. To address this issue, we evaluated several aspects of network resource performance. Despite the general belief, the variation of serverless applications' network performance is quite significant, and the ability to isolate network resource allocation during concurrent execution is rarely provided by service providers. Based on the results presented in this paper, we insist that network resource performance of functional execution models should be more visible and predictable, in order to expand the applications of serverless computing.