Caching Aided Multi-Tenant Serverless Computing

TL;DR

Cold-start latency in multi-tenant serverless platforms is exacerbated by resource-limited edge environments. FaasCamp introduces a multi-tier warm pool with a reclaim pool and an ML-based invoker to approximate Bélády’s optimal eviction, leveraging checkpoint/restore for secure cross-tenant sharing. The system includes a trace-driven training pipeline, an OpenWhisk prototype, and extensive evaluation showing higher warm rates and low overhead, especially for mobile users. These contributions offer a scalable approach to reducing cold starts in resource-constrained, multi-tenant serverless and edge deployments.

Abstract

One key to enabling high-performance serverless computing is to mitigate cold-starts. Current solutions utilize a warm pool to keep function alive: a warm-start can be analogous to a CPU cache-hit. However, modern cache has multiple hierarchies and the last-level cache is shared among cores, whereas the warm pool is limited to a single tenant for security concerns. Also, the warm pool keep-alive policy can be further optimized using cache replacement algorithms. In this paper, we borrow practical optimizations from caching, and design FaasCamp, a caching-aided multi-tenant serverless computing framework. FaasCamp extends the single-tier warm pool into multi-tiers, with a reclaim pool introduced enabling secure function instance sharing among tenants. Also, FaasCamp leverages machine learning to approximate the optimal cache replacement policy to improve the warm rate. We have implemented a prototype and conducted extensive experiments under multiple scenarios. The results show that FaasCamp can outperform existing platforms with minimal overhead.

Figures (12)

• Figure 1: Bélády's cache replacement policy example: Object B with a reuse distance 4 should be evicted.
• Figure 2: Number of cold starts (dots) with respect to tenants.
• Figure 2: Reclaim Pool Performance.
• Figure 3: Average Spike Interval.
• Figure 4: Performance Gap.