SLAM: SLO-Aware Memory Optimization for Serverless Applications

TL;DR

This paper introduces SLAM, a tool that optimizes memory configurations for serverless applications by modeling function relationships and execution times to meet SLOs while minimizing costs.

Contribution

SLAM is the first tool to optimize memory settings for complex serverless workflows using distributed tracing and execution modeling.

Findings

Over 95% of requests meet SLOs with SLAM's configurations.

SLAM reduces costs while maintaining performance.

Effective for multiple real-world AWS Lambda applications.

Abstract

Serverless computing paradigm has become more ingrained into the industry, as it offers a cheap alternative for application development and deployment. This new paradigm has also created new kinds of problems for the developer, who needs to tune memory configurations for balancing cost and performance. Many researchers have addressed the issue of minimizing cost and meeting Service Level Objective (SLO) requirements for a single FaaS function, but there has been a gap for solving the same problem for an application consisting of many FaaS functions, creating complex application workflows. In this work, we designed a tool called SLAM to address the issue. SLAM uses distributed tracing to detect the relationship among the FaaS functions within a serverless application. By modeling each of them, it estimates the execution time for the application at different memory configurations. Using these estimations, SLAM determines the optimal memory configuration for the given serverless application based on the specified SLO requirements and user-specified objectives (minimum cost or minimum execution time). We demonstrate the functionality of SLAM on AWS Lambda by testing on four applications. Our results show that the suggested memory configurations guarantee that more than 95% of requests are completed within the predefined SLOs.