# Benchmarking, Analysis, and Optimization of Serverless Function Snapshots

arXiv: https://arxiv.org/abs/2101.09355 · HTML: https://mlsanity.com/paper/2101.09355/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Jan 2021.

## Built on by (most cited first, 12 of 12)

- [ServerlessLLM: Low-Latency Serverless Inference for Large Language Models](https://mlsanity.com/paper/2401.14351/) (2024)
- [DeltaBox: Scaling Stateful AI Agents with Millisecond-Level Sandbox Checkpoint/Rollback](https://mlsanity.com/paper/2605.22781/) (2026)
- [Serverless Data Science -- Are We There Yet? A Case Study of Model Serving](https://mlsanity.com/paper/2103.02958/) (2021)
- [FedLesScan: Mitigating Stragglers in Serverless Federated Learning](https://mlsanity.com/paper/2211.05739/) (2022)
- [LIFL: A Lightweight, Event-driven Serverless Platform for Federated Learning](https://mlsanity.com/paper/2405.10968/) (2024)
- [PipeBoost: Resilient Pipelined Architecture for Fast Serverless LLM Scaling](https://mlsanity.com/paper/2503.17707/) (2025)
- [Shabari: Delayed Decision-Making for Faster and Efficient Serverless Functions](https://mlsanity.com/paper/2401.08859/) (2024)
- [PerOS: Personalized Self-Adapting Operating Systems in the Cloud](https://mlsanity.com/paper/2404.00057/) (2024)
- [BootSeer: Analyzing and Mitigating Initialization Bottlenecks in Large-Scale LLM Training](https://mlsanity.com/paper/2507.12619/) (2025)
- [Adaptive Serverless Resource Management via Slot-Survival Prediction and Event-Driven Lifecycle Control](https://mlsanity.com/paper/2604.05465/) (2026)
- [Evidence-Aware MapReduce for Forkable Compute](https://mlsanity.com/paper/2607.09689/) (2026)
- [Fast Recovery for LLM Serving via Decoupled Device Memory Lifetime in Dynamo](https://mlsanity.com/paper/2609.25451/) (2026)

## Abstract (arXiv)

Serverless computing has seen rapid adoption due to its high scalability and flexible, pay-as-you-go billing model. In serverless, developers structure their services as a collection of functions, sporadically invoked by various events like clicks. High inter-arrival time variability of function invocations motivates the providers to start new function instances upon each invocation, leading to significant cold-start delays that degrade user experience. To reduce cold-start latency, the industry has turned to snapshotting, whereby an image of a fully-booted function is stored on disk, enabling a faster invocation compared to booting a function from scratch. This work introduces vHive, an open-source framework for serverless experimentation with the goal of enabling researchers to study and innovate across the entire serverless stack. Using vHive, we characterize a state-of-the-art snapshot-based serverless infrastructure, based on industry-leading Containerd orchestration framework and Firecracker hypervisor technologies. We find that the execution time of a function started from a snapshot is 95% higher, on average, than when the same function is memory-resident. We show that the high latency is attributable to frequent page faults as the function's state is brought from disk into guest memory one page at a time. Our analysis further reveals that functions access the same stable working set of pages across different invocations of the same function. By leveraging this insight, we build REAP, a light-weight software mechanism for serverless hosts that records functions' stable working set of guest memory pages and proactively prefetches it from disk into memory. Compared to baseline snapshotting, REAP slashes the cold-start delays by 3.7x, on average.