No abstract

No abstract

Abstract. Subscription adaptations are becoming increasingly important across many content-based publish/subscribe (CPS) applications. In algorithmic high frequency trading, for instance, stock price thresholds that are of interest to a trader change rapidly, and gains directly hinge on the reaction time to relevant fluctuations. The common solution to adapt a subscription consists of a re-subscription, where a new subscription is issued and the superseded one canceled. This is ineffective, leading to missed or duplicate events during the transition. In this paper, we introduce the concept of parametric subscriptions to support subscription adaptations. We propose novel algorithms for updating routing mechanisms effectively and efficiently in classic CPS broker overlay networks. Compared to re-subscriptions, our algorithms significantly improve the reaction time to subscription updates and can sustain higher throughput in the presence of high update rates. We convey our claims through implementations of our algorithms in two CPS systems, and by evaluating them on two different real-world applications.

No abstract

Designing distributed applications around the idiom of events has several benefits including extensibility and scalability. To improve conciseness, safety, and efficiency of corresponding programs, several authors have recently proposed programming languages or language extensions with support for event-based programming.The presence of a dedicated programming language and compilation process offers avenues for program analyses to further improve simplicity, safety, and expressiveness of distributed event-based software. This paper presents three program analyses specifically designed for event-based programs: immutability analysis avoids costly cloning of events in the presence of co-located handlers for same events; guard analysis allows for simple yet expressive subscriptions which can be further simplified and handled efficiently; causality analysis determines causal dependencies among events which are related, allowing unrelated events to be transferred independently for efficiency. We convey the benefits of our approach by empirically evaluating their performance benefits.