TimeBench/SRVN is a prototype toolset for computer-aided design and performance analysis of software, with an emphasis on distributed client-server systems. The performance behaviour of such systems may defy intuition because it involves factors in the software design (such as the partitioning of the functionality and the frequency with which requests will be made to each server) and in the configuration of the distributed system (including replication of services, the distribution of data, and the speed of network access). The novelty of the tool consists in providing support both for developing design specifications and also for performance analysis. The integrated approach avoids the semantic gap between a designer's domain and the performance modeling domain, and assists the designer to explore factors that impact the performance of a design. The performance models are based on the Stochastic Rendezvous Network (SRVN) formalism for client-server systems with synchronous service req...

Over the last decade considerable effort has been put in the development of techniques to assess the performance and the dependability of computer and communication systems in an integrated way. This so-called performability modelling becomes especially useful when the system under study can operate partially, which is for instance the case for fault-tolerant computer systems and distributed systems. Modelling techniques are a fundamental prerequisite for actually doing performability analysis. A prerequisite of a more practical but not less important nature is the availability of software tools to support the modelling techniques and to allow system designers to incorporate the new techniques in the design process of systems. Since performability modelling requires many aspects of a system to be specified, high requirements should be posed on performability modelling tools. Moreover, these tools should be structured such that the models can be specified at a level that is easy to unde...

Introduction The Stochastic Petri Net Package (SPNP) is a versatile modeling tool for the solution of Stochastic Petri Nets (SPN) models. The SPN models are described in the input language for SPNP called CSPL (C-based SPN Language). The CSPL is an extension of the ANSI C programming language [16] with additional constructs to facilitate the easy description of SPN models. The full power and generality of C is available, but a working knowledge of C is sufficient to use SPNP effectively. The SPN models specified to SPNP are actually "SPN Reward Models" or Stochastic Reward Nets (SRNs) [9, 10] which are based on the "Markov Reward Model" paradigm [18, 37]. This provides a powerful modeling environment for the analysis of: ffl Dependability (Reliability, Availability, Safety). ffl Performance. ffl Performability. Several important Petri net constructs like marking dependency, variable cardinality arc and enabling functions [9] facil

The Layered System Generator is used to create synthetic distributed systems of tasks with client-server style (RPC) interactions, representing a wide range of software architectures and workload patterns. A synthetic task system can be used to generate network and workstation traffic which represents the load from a planned software system, so one can observe its probable performance when run on the target network, or its probable impact on other existing applications. It can be used to evaluate the planned software design, or the target network's capability, or both combined. Using LSG, tests were made with systems of up to 39 tasks on a UNIX network, to investigate the performance changes that occur when a small task system is scaled up in size. The performance recorded across the range of experiments was also compared with predictions made by an analytic performance model. The errors were found to be small provided an allowance is made for workstation daemons and similar load components.

: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : xiii 1. INTRODUCTION : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 1 2. BACKGROUND AND RELATED WORK : : : : : : : : : : : : : : : : : : 7 2.1 Basic Concepts in Heterogeneous Computing : : : : : : : : : : : : : : 7 2.2 Mapping, Matching, and Scheduling : : : : : : : : : : : : : : : : : : : 10 2.3 Characterizing Matching and Scheduling for HC Systems : : : : : : : 11 2.4 Examples of Techniques and Formulations for Matching and Scheduling for HC Systems : : : : : : : : : : : : : : 12 2.4.1 Block-based SIMD/SPMD mode selection technique and its extension : : : : : : : : : : : : : : : : : : : : : : : : : 12 2.4.2 Optimal selection theory and its extensions : : : : : : : : : : : 18 2.4.3 Other formulations and solution techniques : : : : : : : : : : : 23 2.5 Summary and Motivation for Proposed Work : : : : : : : : : : : : : : 25 3. PROBABILITY DEFINITIONS AND NOTATIONS USED : : : : : : : : 27 3.1 Definiti...

With advances in distributed platforms, distributed application systems on heterogeneous platforms have become a reality. These systems are complex, with software and hardware components that are distributed geographically, sometimes globally, yet they combine to provide a single processing model. Deciding how to distribute software and hardware resources for these systems is a major challenge and is influenced by factors such as security, reliability, manageability, and performance. In this paper, we focus on the performance behaviour of a design for a CORBA-based Distributed Network Management Application. Once the design is established and the system's development begins, performance problems resulting from a poor design can be difficult and expensive to fix. We built both simulation and analytic models, and used the models to help recognize such problems early. We discuss the proposed system architecture, simulation and analytic models for the application, and extensions to the analytic models needed to support features present in the design.

In this paper we present stochastic Petri nets (SPNs) that can be used for the evaluation of polling mechanisms. Polling mechanisms (or systems) appear in many forms in computer-communication systems: the well-known token-ring and token-bus network access schemes such as present in IEEE P802.4/5 and in FDDI, and the scheduling mechanisms in switching fabrics, e.g., for ATM systems, operate along the lines of polling systems. Polling systems have been studied for many years now, and many analytical techniques have been developed to study them. It seems, however, that a number of system aspects can not be covered adequately by such analytical approaches. Most notably are time-dependent polling variants where the amount of service a station receives per visit is time-limited, load-dependent polling strategies where the ordering of station visits is dependent on the loading of the stations, as well as non-Poisson arrival processes. Therefore, we present SPN-based models that allow us to cope with these system aspects. The two major problems that appear when taking the SPN approach are the size of the underlying CTMC and the use of non-exponential timing. The latter problem is not really addressed in this paper; rather it is circumvented by employing the well-known method of stages, thus even worsening the state-space size problem. The first problem is coped with, by presenting two decomposition approaches and by presenting a subclass of SPNs that allows for an efficient matrix-geometric solution, thus avoiding the explicit generation of the overall state space.

this paper we present a distributed monitor called ANSAmon, that we developed for the ANSAware platform. It is realized according to a master-slave architecture and is a distributed application itself. Monitoring experiments can be configured in advance, whereas analysis is done after collecting all events. Global times will be estimated by a new method improving existing ones. Applying ANSAmon we investigated the ANSA REX protocol and the ANSA trader. Important conclusions for the efficient use of ANSAware for distributed applications can be deduced from the results obtained. Keyword Codes: C.2.4; C.4;D.4.4 Keywords: Distributed Systems, Performance of Systems, Communications Management 1. INTRODUCTION The ongoing migration from large mainframe computers to networks of workstations derives new requirements to application programmers. Whereas the former systems allow monolitic programs to be developed, modern systems are organized in a client-server fashion. As a consequence, a means for programming inter-workstation cooperation is necessary. The most common method is the remote procedure call (RPC), see [BiNe84]. Several distributed computing platforms offering appropriate services have been developed in the last years. Examples are the distributed computing environment (DCE) of the Open Software Foundation [DCE], the common object request broker architecture (CORBA) of the Object Management Group [OMG] or the ANSAware (Advanced Network Systems Architecture) of the AMP [ANSA]. Besides these commericial products, ISO developes a reference model for Open Distributed Processing (ODP) [ODP p2], [ODP p3]. It consists of concepts and their * This work was partially supported by the Deutsche Forschungsgemeinschaft under Grant No. Sp 230/8-1 formal interpretation for distrib...

Abstract—With the increasing popularity of wireless communication systems, customers are expecting the same level of service, availability, and performance from the wireless communication networks as the traditional wire-line networks. Traditional pure performance model that ignores failure and recovery but considers resource contention generally overestimates the system’s ability to perform a certain job. On the other hand, pure availability analysis tends to be too conservative since performance considerations are not taken into account. To obtain realistic composite performance and availability measures, one should consider performance changes that are associated with failure recovery behavior. In this paper, a brief review is first given over the advances in composite performance and availability analysis. Thereafter, three techniques for composite performance and availability analysis are discussed in detail through a queueing system in a wireless communication network. Numerical results show that an approximate model based on a framework originally proposed by Bobbio and Trivedi (BT approach) provides remarkably accurate predictions on system performance. Index Terms—Availability, channel allocation, failure recovery, performability, stochastic reward net (SRN), wireless communication systems. I.

With advances in distributed platforms, distributed application systems on heterogeneous platforms have become a reality. These systems are complex, with software and hardware components that are distributed geographically, sometimes globally, yet they combine to provide a single processing model. Deciding how to distribute software and hardware resources for these systems is a major challenge and is influenced by factors such as security, reliability, manageability, and performance. In this paper, we focus on the performance behaviour of a design for a CORBA-based Distributed Network Management Application. Once the design is established and the system's development begins, performance problems resulting from a poor design can be difficult and expensive to fix. We built both simulation and analytic models, and used the models to help recognize such problems early. We discuss the proposed system architecture, simulation and analytic models for the application, and extensions to the analytic models needed to support features present in the design.