Berlin, Techn. University, Diss., 2007.

Berlin, Techn. Univ., Diss., 2005. / Computerdatei im Fernzugriff.

Many large business processes are inherently distributed, spanning multiple organizations, administrative domains, and geographic locations. To support such applications, this thesis develops a flexible and distributed platform to develop, execute, and monitor business processes. The solutions utilize a distributed content-based publish/subscribe overlay that is extended with support for mobile clients and client interest churn. Over this layer, a distributed execution engine uses events to coordinate the execution of the process, and dynamically redeploys activities in the process in order to minimize a user-specified cost function and preserve service level agreements (SLAs). Finally, a management layer allows users to find and automatically compose services available across a distributed set of service registries, and monitor processes for SLA violations. Evaluations show that the distributed execution engine can scale better than alternate architectures, exhibiting over 60% improvements in execution time in one experiment. As well the system can dynamically redeploy processes to reflect changing workload conditions and SLAs, saving up to 90% of the process messaging overhead of a static deployment.

Many large business processes are inherently distributed, spanning multiple organizations, administrative domains, and geographic locations. To support such applications, this thesis develops a flexible and distributed platform to develop, execute, and monitor business processes. The solutions utilize a distributed content-based publish/subscribe overlay that is extended with support for mobile clients and client interest churn. Over this layer, a distributed execution engine uses events to coordinate the execution of the process, and dynamically redeploys activities in the process in order to minimize a user-specified cost function and preserve service level agreements (SLAs). Finally, a management layer allows users to find and automatically compose services available across a distributed set of service registries, and monitor processes for SLA violations. Evaluations show that the distributed execution engine can scale better than alternate architectures, exhibiting over 60% improvements in execution time in one experiment. As well the system can dynamically redeploy processes to reflect changing workload conditions and SLAs, saving up to 90% of the process messaging overhead of a static deployment.

In the publish/subscribe paradigm, information providers disseminate publications to all consumers who have expressed interest by registering subscriptions. This paradigm has found wide-spread applications, ranging from selective information dissemination to network management. However, all existing publish/subscribe systems cannot capture uncertainty inherent to the information in either subscriptions or publications. In many situations the large number of data sources exhibit various kinds of uncertainties. Examples of imprecision include: exact knowledge to either specify subscriptions or publications is not available; the match between a subscription and a publication with uncertain data is approximate; the constraints used to define a match is not only content based, but also take the semantic information into consideration. All these kinds of uncertainties have not received much attention in the context of publish/subscribe systems. In this thesis, we propose new publish/subscribe models to express uncertainties and semantics in publications and subscriptions, along with the matching semantics for each model. We also develop efficient algorithms to perform filtering for our models so that it can be applied to process the rapidly increasing information on the Internet. A thorough experimental evaluation is presented to demonstrate that the proposed systems can offer scalability to large number of subscribers and high publishing rates.

In the era of mobile cloud computing, the consumption of virtualized software and Web-based services from super-back-end infrastructure using smartphones and tablets is gaining much research attention from both the industry and academia. Nowadays, these mobile devices generate and access multimedia data hosted in social media and other sources in order to enhance the users’ multimedia experience. However, multimedia data is unstructured which can lead to challenges with data synchronization between these mobile devices and the cloud computing back-end. The issue with data synchronization is further fueled by the fact that mobile devices can experience intermittent connectivity losses due to unstable wireless bandwidths. While previous works proposed Simple Object Access Protocol (SOAP) -based middleware for the Web services’ synchronization, this approach is not efficient in a mobile environment because the SOAP protocol is verbose. Thus, the Representational State Transfer (REST) standard has been proposed recently to model the Web services since it is lightweight. This thesis proposes a novel approach for implementing a REST-based mobile Web Service for multimedia file sharing that utilizes a channel-based publish/subscribe communication scheme to synchronize smartphone or tablet-hosted NoSQL databases with a cloud-hosted NoSQL database. This thesis evaluates the synchronicity and the scalability of a prototype system that was implemented according to this approach. Also, this thesis assesses the overhead of the middleware component of the system.

In the publish/subscribe paradigm, information providers disseminate publications to all consumers who have expressed interest by registering subscriptions. This paradigm has found wide-spread applications, ranging from selective information dissemination to network management. However, all existing publish/subscribe systems cannot capture uncertainty inherent to the information in either subscriptions or publications. In many situations the large number of data sources exhibit various kinds of uncertainties. Examples of imprecision include: exact knowledge to either specify subscriptions or publications is not available; the match between a subscription and a publication with uncertain data is approximate; the constraints used to define a match is not only content based, but also take the semantic information into consideration. All these kinds of uncertainties have not received much attention in the context of publish/subscribe systems. In this thesis, we propose new publish/subscribe models to express uncertainties and semantics in publications and subscriptions, along with the matching semantics for each model. We also develop efficient algorithms to perform filtering for our models so that it can be applied to process the rapidly increasing information on the Internet. A thorough experimental evaluation is presented to demonstrate that the proposed systems can offer scalability to large number of subscribers and high publishing rates.

Techn. University, Diss., 2004--Darmstadt.

In this study a comparison of the publish-subscribe communication protocols OPS, MQTT and DDS were conducted. The implementation used for DDS was fastDDS and the client library Paho Eclipse MQTT C++ was used for MQTT with the broker implementation Eclipse Mosquitto. The goal was to see which out of these protocols performs best in a peer-to-peer communication scenario when it comes to latency and throughput with varying payload size. Two experiments were carried out. The first experiment measures the round-trip-time of a message and is realized by a ping-pong application. The latency was then calculated by taking half of the round-trip-time. The second experiment was publishing messages from one client and being retrieved by another and counting the amount of messages being retrieved in one second. To get bytes per second the payload size was multiplied by the amount of messages retrieved by the subscriber. The result of the first experiment showed that OPS with the underlying transport protocol TCP had the most favorable result. For the second experiment OPS with UDP as the transport protocol had the highest throughput in most cases however for the highest payload used in the experiment fastDDS with UDP had the highest throughput. However, the results gave substantial differences between the protocols in regards to throughput which indicate that the configuration might not be optimal for a fair comparison. Further research of this is recommended before making a final conclusion in regards to the overall results. / I denna studie genomfördes en jämförelse av kommunikationsprotokollen OPS, MQTT och DDS. Implementeringen som användes för DDS var fastDDS och klientbiblioteket Paho Eclipse MQTT C++ användes för MQTT med Eclipse Mosquitto som broker. Målet var att se vilket av dessa protokoll som fungerar bäst i ett peer-to-peer-kommunikationsscenario när det gäller fördröjningstid och throughput med varierande payloadstorlek. Två experiment utfördes. Det första experimentet mäter round-trip-time för ett meddelande och realiseras av en ping-pong-applikation. Fördröjningstiden beräknades sedan genom att ta hälften av round-trip-time. Det andra experimentet var att publicera meddelanden från en klient och mottas av en annan och räkna mängden meddelanden som mottas på en sekund. För att få byte per sekund multiplicerades payload storleken med mängden meddelanden som togs emot av prenumeranten. Resultatet av det första experimentet visade att OPS med det underliggande transportprotokollet TCP hade det mest gynnsamma resultatet. För det andra experimentet hade OPS med UDP som transportprotokoll den högsta throughput i de flesta fall, men för den största payloadstorleken som användes i experimentet hade fastDDS med UDP den högsta. Resultaten gav dock betydande skillnader mellan protokollen när det gäller throughput som indikerar att konfigurationen kanske inte är optimal för en rättvis jämförelse. Ytterligare forskning om detta rekommenderas för att utreda orsaken till dessa skillnader och därefter kunna skapa konfigurationer som ger rättvisa förutsättningar för en optimal jämförelse.

This thesis work has aimed to implement a robust communication system for military aircrafts produced by Saab. A big part of the thesis focuses on a comparison study of different possible designs and their impact on the system. From the comparison study a decentralized, Peer to Peer, Publish/Subscribe system was selected for implementation. All publications are sent directly from a publisher to the subscribers without any intermediate forwarding node. This has shown good results in comparison to the previous centralized solution where all data had to pass through a single server node. The new system has one master node that is responsible for registrations of both publishers and subscribers and distribute any necessary information to affected clients. This thesis has shown that the Single Point of Failure that was one of the main issues on the previous design has been removed and the performance of the system has increased as well. The Round Trip Time of a set of messages has shown to be improved up to 70.78%.