Efficient Serverless Cold Start: Reducing Library Loading Overhead by Profile-guided Optimization

2025-04-27

AI Summary

High cold-start latency in serverless computing—particularly due to unnecessary library loading during initialization—remains a critical bottleneck. To address this, we propose a runtime profiling-driven, workload-evolution-aware library trimming method. Unlike static analysis, our approach constructs fine-grained library usage profiles via statistical sampling and call-path analysis, then applies automated code refactoring to dynamically and precisely identify and eliminate infrequently or never-used libraries. The framework integrates seamlessly into CI/CD pipelines and supports adaptive, continuous optimization. Evaluated on multiple benchmarks and real-world applications, it achieves up to 56.5% reduction in initialization latency (2.30× speedup), 55.8% improvement in end-to-end latency (2.26×), and 33.1% lower memory footprint (1.51×). To the best of our knowledge, this is the first work to realize closed-loop, behavior-driven library minimization for serverless functions.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Algorithm Configuration
• Computer Vision: Learning & Optimization for CV

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Security and Privacy: Tracking, profiling, and countermeasures against them

Abstract

Serverless computing abstracts away server management, enabling automatic scaling, efficient resource utilization, and cost-effective pricing models. However, despite these advantages, it faces the significant challenge of cold-start latency, adversely impacting end-to-end performance. Our study shows that many serverless functions initialize libraries that are rarely or never used under typical workloads, thus introducing unnecessary overhead. Although existing static analysis techniques can identify unreachable libraries, they fail to address workload-dependent inefficiencies, resulting in limited performance improvements. To overcome these limitations, we present SLIMSTART, a profile-guided optimization tool designed to identify and mitigate inefficient library usage patterns in serverless applications. By leveraging statistical sampling and call-path profiling, SLIMSTART collects runtime library usage data, generates detailed optimization reports, and applies automated code transformations to reduce cold-start overhead. Furthermore, SLIMSTART integrates seamlessly into CI/CD pipelines, enabling adaptive monitoring and continuous optimizations tailored to evolving workloads. Through extensive evaluation across three benchmark suites and four real-world serverless applications, SLIMSTART achieves up to a 2.30X speedup in initialization latency, a 2.26X improvement in end-to-end latency, and a 1.51X reduction in memory usage, demonstrating its effectiveness in addressing cold-start inefficiencies and optimizing resource utilization.

Problem

• Reducing cold-start latency in serverless computing
• Optimizing library usage to minimize initialization overhead
• Improving resource utilization and performance in serverless applications

Innovation

• Profile-guided optimization for library usage
• Statistical sampling and call-path profiling
• Automated code transformations for cold-start