Robert Henjes

German-lab experimental facility

The G-Lab project aims to investigate concepts and technologies for future networks in a practical manner. Thus G-Lab consists of two major fields of activities: research studies of future network components and the design and setup of experimental facilities. Both is controlled by the same community to ensure, that the experimental facility fits to the demand of researchers. Researchers gain access to virtualized resources or may gain exclusive access to resource if necessary. We present the current setup of the experimental facility, describing the available hardware, management of the platform, the utilization of the Planet-Lab software and the user management.

Throughput Performance of Java Messaging Services Using WebsphereMQ

The Java messaging service (JMS) is a middlewareoriented messaging technology working according to the publish/ subscribe principle. If subscribers install filter rules on the JMS server, JMS can be used as a message routing platform, but it is not clear whether its message throughput is sufficiently high to support large-scale systems. In this paper we investigate the capacity of the Websphere Message Queue JMS server. In contrast to other studies, we focus on the message throughput in the presence of filters and show that filtering reduces the performance significantly. We also present a model that describes the service time for a single message depending on the number of installed filters and validate it by measurements. This model helps to calculate the system throughput for specific application scenarios.

Analysis of the Message Waiting Time for the FioranoMQ JMS Server

The Java messaging service (JMS) is a means to organize communication among distributed applications according to the publish/subscribe principle. If the subscribers install filter rules on the JMS server, JMS can be used as a message routing platform, but it is not clear whether its message throughput is sufficiently high to support large-scale systems. We perform measurements for the FioranoMQ JMS server and derive a simple model for its message processing time that takes message filters and the message replication grade into account. Then, we analyze the JMS server capacity and the message waiting time for various application scenarios. We show that the message waiting time is not an issue as long as the server throughput is sufficiently high. Finally, we assess the capacity of two different distributed JMS architectures whose objective is to increase the capacity of the JMS beyond the limit of a single server.

On the stability of chord-based P2P systems

The current generation of P2P networks is intended to provide cost-effective alternatives to the traditional client-server architecture. The main goal is to store and retrieve data in a decentralized manner. The challenge in doing so consists in creating a stable overlay network that allows for fast and efficient searches. In this paper we consider the Chord P2P algorithm in this context and analyze its stability and efficiency with stochastic methods. We present realistic probabilities for a disconnection and investigate the corresponding scalability.

Throughput performance of popular JMS servers

The Java Messaging Service (JMS) facilitates communication among distributed software components according to the publish/subscribe principle. If the subscribers install filter rules on the JMS server, JMS can be used as a message routing platform, but it is not clear whether its message throughput is sufficiently high to support large-scale systems. In this paper, we investigate the capacity of three high performance JMS server implementations: FioranoMQ, SunMQ, and WebsphereMQ. In contrast to other studies, we focus on the message throughput in the presence of filters and show that filtering reduces the performance significantly. We present models for the message processing time of each server and validate them by measurement.

A scalable algorithm to monitor chord-based p2p systems at runtime

Peer-to-peer (p2p) systems are a highly decentralized, fault tolerant, and cost effective alternative to the classic client-server architecture. Yet companies hesitate to use p2p algorithms to build new applications. Due to the decentralized nature of such a p2p system the carrier does not know anything about the current size, performance, and stability of its application. In this paper we present an entirely distributed and scalable algorithm to monitor a running p2p network. The snapshot of the system enables a telecommunication carrier to gather information about the current performance parameters of the running system as well as to react to discovered errors.

Throughput Performance of the ActiveMQ JMS Server

Communication among distributed software components according to the publish/subscribe principle is facilitated by the Java messaging service (JMS). JMS can be used as a message routing platform if the subscribers install filter rules on the JMS server. However, it is not clear whether its message throughput is sufficient to support large-scale systems. In this paper, we investigate the capacity of the high performance JMS server implementation ActiveMQ. In contrast to other studies, we focus on the message throughput in the presence of filters and show that filtering reduces the performance significantly. We present a model for the message processing time at the server and validate it by measurements. This model takes the number of installed filters and the replication grade of the messages into account and predicts the overall message throughput for specific application scenarios.

Implementation of Adaptive Control for P2P Overlays

Peer-to-peer networking enjoys euphoric support and fierce resistance simultaneously, and for the same reasons. It presents a model where decentralization and lack of structure, hierarchy and control are promoted. Although significant research is carried out to tackle individual issues arising from that paradigm, there has been no obvious approach for evening out differences on a more general basis. In this paper we introduce a framework and provide implementation techniques for such an approach. The framework aims at integrating partial techniques that solve individual problems and has been designed for flexibility. The integrated approach we are proposing includes forming and maintaining of peer-to-peer overlays, controlling the underlying topology being formed, limiting the signaling traffic being generated and optimizing the payload traffic.

Gamma-approximation for the waiting time distribution function of the M/G/1-/spl infin/ queue

In this paper we propose to approximate the waiting time distribution function (DF) of the waiting customers in a M/G/1-infin queuing system by a gamma-distribution whose parameters alpha and beta are set by means of the first and second moment of the waiting time that are obtained from the Takacs recursion formula. Discrete-time analysis (DTA) is another approximation approach for the same objective. We show first for analytically feasible special cases that DTA is very accurate and then we use it to validate the accuracy of the new gamma-approximation for wide parameter ranges regarding the service time distribution. We show that the gamma-approximation respects well even the third moment of the service time distribution. As the new approach is very simple and fast, it may be used by engineers with only little background in queuing theory to calculate quantiles for real-time control loops in technical systems

Impact of complex filters on the message throughput of the activeMQ JMS server

In this paper we investigate the maximum message throughput of the ActiveMQ server in different application scenarios. We use this throughput as a performance criterion. It depends heavily on the installed filters and the message replication grade. In previous work, we have presented measurement results and an analytical model for simple filters. This work extends these studies towards more complex configuration options. It provides measurement results and analytical performance models for complex AND-, OR-, and IN-filters. The results are useful to understand the performance of JMS servers and help to dimension large distributed JMS-based systems.