Network-Aware Operator Placement for Stream-Processing Systems

TLDR

Distributed stream‑processing systems push query operators onto network nodes, but manual placement is difficult because network and node conditions fluctuate and streams can interact, creating opportunities for reuse and repositioning. This paper introduces the stream‑based overlay network (SBON), a layer that manages operator placement between a stream‑processing system and the physical network. SBON’s design uses a cost‑space abstraction of the network and active streams to enable decentralized, large‑scale multi‑query optimization, and is evaluated through simulation, PlanetLab experiments, and integration with the Borealis engine. Results demonstrate that SBON consistently improves network utilization, reduces stream latency, and supports dynamic optimization with minimal engineering effort.

Abstract

To use their pool of resources efficiently, distributed stream-processing systems push query operators to nodes within the network. Currently, these operators, ranging from simple filters to custom business logic, are placed manually at intermediate nodes along the transmission path to meet application-specific performance goals. Determining placement locations is challenging because network and node conditions change over time and because streams may interact with each other, opening venues for reuse and repositioning of operators. This paper describes a stream-based overlay network (SBON), a layer between a stream-processing system and the physical network that manages operator placement for stream-processing systems. Our design is based on a cost space, an abstract representation of the network and on-going streams, which permits decentralized, large-scale multi-query optimization decisions. We present an evaluation of the SBON approach through simulation, experiments on PlanetLab, and an integration with Borealis, an existing stream-processing engine. Our results show that an SBON consistently improves network utilization, provides low stream latency, and enables dynamic optimization at low engineering cost.