Benchmark reveals autoscaling matters more than placement in cloud edge computing
ContinuumBench: Benchmarking Joint Autoscaling and Placement Across Evaluation Regimes in the Cloud-Edge Continuum
Distributed, Parallel, and Cluster Computing
Summary
Keeping online applications running smoothly often means deciding where to put tasks and how many copies to make. The authors created a testing method called ContinuumBench to fairly compare these decisions under different conditions. They found that having enough resource capacity and smart autoscaling is more important for meeting deadlines than the exact placement of tasks, unless tasks can be moved easily at no cost. Also, how you count late or unfinished work changes which methods look better. This work helps clarify what really matters in managing cloud and edge computing resources.
autoscalingservice placementcloud-edge computinglatencyresource managementbenchmarkworkloaddeadline missessimulationcapacity
Authors
Lanpei Li, Antonino Vaccarella, Vincenzo Lomonaco, Massimo Coppola
Abstract
Cloud-edge controllers coordinate service placement, replica scaling, and resource pre-warming to keep end-to-end latency within application deadlines. But evaluations often obscure the source of a reported gain: placement and scaling are studied separately; workload, connectivity, and calibration assumptions remain implicit; and metrics over completed tasks hide unfinished work. We present ContinuumBench, a benchmark that controls these factors. Its completion-aware accounting treats late, unfinished, and discarded tasks as deadline misses. A common protocol compares placement-only and scale-capable controllers under declared regimes and stressors. Built on the ECLYPSE simulator, ContinuumBench adds arrivals, worker elasticity, intermittent transport, buffering, and failures to close the control loop. We evaluate nine controllers across four scenarios and two regimes. The studied regimes are capacity-bound: elastic capacity, not placement sophistication, drives completion, and once capacity suffices, the choice of autoscaling policy decides how much of that work arrives on time. Placement re-planning has no measurable effect without relocation, while cost-free migration defines the observed exception. Consequently, scale-capable controllers approach an over-provisioned reference while placement-only controllers degrade with load; and placement quality separates controllers only once capacity is exhausted. Finally, the accounting choice itself changes the reported result: completion-only and completion-aware scoring can rank controllers differently.