Design and evaluation of a Pub/Sub service in the cloud

Abstract

Pub/Sub services in the cloud are intended to achieve high performance and high scalability. The challenge comes from how to deal with the potential wide variation in workloads. The paper constructs a content-based Pub/Sub service named OPS4Cloud on the top of cloud infrastructure from scratch. OPS4Cloud manages its brokers by regions and presents a VM-based region adjustment strategy. It builds two-level indexes at brokers to speed up the routing of subscriptions, events and advertisements on the basis of channelization routing. It also designs a mechanism for subscription persistence, accelerating the load transfer. The paper conducts extensive experiments on OPS4Cloud, especially in comparison with the migrated Pub/Sub systems. The experimental data shows that OPS4Cloud is vastly superior to the migrated systems in terms of system throughput and event response time, and it also tends towards a load-balancing state among brokers while running continuously.

Elastic Scaling of a High-Throughput Content-Based Publish/Subscribe Engine

“…Finally, BlueDove supports only scale out, therefore, unlike E-STREAMHUB, it cannot be described as fully elastic. Fang et al [14] follow a similar approach to BlueDove and only scales out.…”

“…We therefore use the ASPE [11] encrypted filtering scheme, and workloads of pre-encrypted subscriptions and publications. While the performance of plain-text filtering may depend on the characteristics of the workload, such as the possibility to leverage containment between subscriptions [4], [7], [9], [10], [13], [14], [22], encrypted filtering such as with ASPE requires filtering each incoming publication against all stored subscriptions. Each individual filtering operation cost is quadratic in number of attributes (O(d 2 )).…”

Reliable messaging to millions of users with migratorydata

“…Existing pub/sub systems can be classified in two categories [7]: content-based [1,8,10] and topic-based pub/sub. We focus in this paper on the topic-based model.…”

Análise Dinâmica do Comportamento de Filas de Mensagens para o Aumento do Paralelismo de Consumo

“…As atribuições feitas pelos publishers aos subscribers podem gerar assimetrias de distribuição dos dados [Tran et al 2011][Li et al 2011][Fang et al 2011. Essas assi-metrias também são comuns em sistemas com múltiplas filas de mensagens e múltiplos consumidores [Tran et al 2011][Li et al 2011.…”

“…As abordagens [Guo et al 2012][Imai et al 2012][Leitner et al 2012] [Ma et al 2010][Sugiki and Kato 2011][Tran et al 2011 citam que a elasticidade deve ser alcançada de forma transparente e automática. Também foi abordado o uso do padrão de projeto publisher/subscriber [Fang et al 2011][Li et al 2011][Tran et al 2011] para alcançar a elasticidade e o desacoplamento de produtores e consumidores.…”