Analysis of Timing Constraints in Heterogeneous Middleware Interactions

Abstract

With the emergence of Future Internet applications that connect web services, sensor-actuator networks and service feeds, scalability and heterogeneity support of interaction paradigms are of critical importance. Heterogeneous interactions can be abstractly represented by client-service, publish-subscribe and tuple space middleware connectors that are interconnected via bridging mechanisms providing interoperability among the services. In this paper, we make use of the eXtensible Service Bus (XSB), proposed in the CHOReOS project as the connector enabling interoperability among heterogeneous choreography participants. XSB models transactions among peers through generic post and get operations that represent peer behavior with varying time/space coupling. Nevertheless, the heterogeneous lease and timeout constraints of these operations severely affect latency and success rates of transactions. By precisely studying the related timing thresholds using timed automata models, we verify conditions for successful transactions with XSB connectors. Furthermore, we statistically analyze through simulations, the effect of varying lease and timeout periods to ensure higher probabilities of successful transactions. Simulation experiments are compared with experiments run on the XSB implementation testbed to evaluate the accuracy of results. This work can provide application developers with precise design time information when setting these timing thresholds in order to ensure accurate runtime behavior.

Leveraging the Service Bus Paradigm for Computer-Mediated Social Communication Interoperability

“…Communication protocols underlying distributed systems may be classified along the following coupling dimensions [15], [16], [20]:…”

“…This is illustrated in We then need to study the extent to which different types of social interaction may be reconciled and when it is appropriate to synthesize the corresponding communication protocol adaptation. To do so, we build upon the eXtensible Service Bus (XSB) [15], [16] that is an approach to reconcile the middleware protocols run by networked systems across the various coupling dimensions (i.e., space, time, synchronization). This leads us to introduce the Social communication bus paradigm.…”

“…The eXtensible Service Bus (XSB) [15], [16] defines a connector that abstracts and unifies three interaction paradigms found in distributed computing systems: client-server, a common paradigm for Web services where a client communicates directly with a server; publish-subscribe, a paradigm for content broadcasting; and tuple-space, a paradigm for sharing data with multiple users who can read and modify that data. XSB is implemented as a common bus protocol that enables interoperability among services employing heterogeneous interactions following one of these computing paradigms.…”

“…We specifically build on the XSB bus [15], [16] that supports interoperability across interaction paradigms as opposed to interoperability across heterogeneous middleware protocols implementing the same paradigm. The proposed bus architecture features the traditional concepts of bus protocols and binding components, but those are customized for the sake of social interactions whose couplings differ along the social and presence dimensions.…”

Towards building real-time, convenient route recommendation system for public transit

“…Async PS interactions will be supported by using the RabbitMQ 10 protocol (commonly used for mobile applications). Subsequently, our previous work on the analysis of timing constraints in heterogeneous middleware interactions [17] will enable our middleware to deal with intermittent network connectivity, and distinguish among the supported interactions to improve the message delays vs. delivery success rates. Moreover, estimated message delays derived using our queueing analysis will enable our middleware to apply appropriate timing parameters.…”