FaaSpot runs serverless workflows cheaply and reliably on spot instances
FaaSpot: Fault-Tolerant Serverless Workflow Execution on Spot Instances
Distributed, Parallel, and Cluster Computing
Summary
Running complex workflows on cloud servers can be expensive and slow, especially when using reliable on-demand machines. The authors found that using cheaper 'spot' machines often causes interruptions that break workflows, especially at critical points where many parts depend on one task. They created FaaSpot, a system that looks at the workflow structure ahead of time to identify these risky tasks and protect them using special techniques. This lets every task run on spot machines without fallback, saving money and finishing faster than existing methods.
What this means in practice
- •For cloud infrastructure teams: Deploy cost-effective serverless workflow platforms that run entirely on spot instances without fallback to on-demand machines by using FaaSpot’s pre-execution task analysis.
- •For cloud cost management teams: Reduce running costs and workflow latency for complex serverless applications by integrating FaaSpot’s survival-aware checkpointing and scaling methods for spot instance usage.
Authors
Bandhan Patel, Sushruth Harsha, Kwangsung Oh, In Kee Kim
Abstract
Serverless computing is increasingly used to run complex DAG workflows across diverse applications, and individual function execution times have grown to support them, pushing workflow makespans to hours. Open-source serverless platforms, deployed on a pool of VMs, typically use on-demand instances, and their cost increases with the makespan. Spot instances are far cheaper, but a preemption is likely before the workflow finishes. A preemption at a fan-out node or a task feeding a fan-in node stalls every parallel branch, cascading a single preemption into a workflow-wide failure. Prior systems address this challenge by migrating high-impact tasks to reliable, non-spot resources, sacrificing spot's cost advantage. We present FaaSpot, a serverless workflow execution system that runs every workflow task on spot instances with no on-demand fallback. Our key insight is that cascade-prone positions are determined by the DAG's topology and can be identified before execution. Therefore, FaaSpot analyzes the workflow DAG once at submission time and classifies each task by the impact of its preemption. Selective checkpointing, survival-aware scaling, and lineage-based retry then act on this single classification, with no runtime coordination. We evaluate FaaSpot on a trace-driven spot emulator with eight DAG workflows against four state-of-the-art baselines. FaaSpot completes 89.03% of runs at $0.018 per successful run, 21pp (percentage points) above the average baseline success rate and at a lower cost per success than every baseline, while reducing mean latency by 2.3x.