This paper describes an accurate and efficient method to model and predict the performance of distributed/parallelsystems. Various performance measures, such as the expected user response time, the system throughput ...
详细信息
This paper describes an accurate and efficient method to model and predict the performance of distributed/parallelsystems. Various performance measures, such as the expected user response time, the system throughput and the average server utilization, can be easily estimated using this method. The methodology is based on known product form queueing network methods, with some additional approximations. The method is illustrated by evaluating performance of a multi-client multi-server distributed system. A system model is constructed and mapped to a probabilistic queueing network model which is used to predict its behavior. The effects of user think time and various design parameters on the performance of the system are investigated by both the analytical method and computer simulation. The accuracy of the former is verified. The methodology is applied to identify the bottleneck server and to establish proper balance between clients and servers in distributed/parallelsystems.
This paper introduces a novel diagnosis approach, using game theory, to solve the comparison-based system-level fault identification problem in distributed and parallel systems based on the asymmetric comparison model...
详细信息
This paper introduces a novel diagnosis approach, using game theory, to solve the comparison-based system-level fault identification problem in distributed and parallel systems based on the asymmetric comparison model. Under this diagnosis model tasks are assigned to pairs of nodes and the results of executing these tasks are compared. Using the agreements and disagreements among the nodes' outputs, i.e. the input syndrome, the fault diagnosis algorithm identifies the fault status of the system's nodes, under the assumption that at most t of these nodes can permanently fail simultaneously. Since the introduction of the comparison model, significant progress has been made in both theory and practice associated with the original model and its offshoots. Nevertheless, the problem of efficiently identifying the set of faulty nodes when not all the comparison outcomes are available to the fault identification algorithm prior to initiating the diagnosis phase, i.e. partial syndromes, remains an outstanding research issue. In this paper, we first show how game theory can be adapted to solve the fault diagnosis problem by maximising the payoffs of all players (nodes). We then demonstrate, using results from a thorough simulation, the effectiveness of this approach in solving the fault identification problem using partial syndromes from randomly generated diagnosable systems of different sizes and under various fault scenarios. We have considered large diagnosable systems, and we have experimented extreme faulty situations by simulating all possible fault sets even those that are less likely to occur in practice. Over all the extensive simulations we have conducted, the new game-theory-based diagnosis algorithm performed very well and provided good diagnosis results, in terms of correctness, latency, and scalability, making it a viable addition or alternative to existing diagnosis algorithms.
This paper introduces a novel diagnosis approach, using game theory, to solve the comparison-based systemlevel fault identification problem in distributed and parallel systems based on the asymmetric comparison model....
详细信息
ISBN:
(纸本)9781509043149
This paper introduces a novel diagnosis approach, using game theory, to solve the comparison-based systemlevel fault identification problem in distributed and parallel systems based on the asymmetric comparison model. Under this diagnosis model tasks are assigned to pairs of nodes and the results of executing these tasks are compared. Using the agreements and disagreements among the nodes' outputs, i.e., the input syndrome, the fault diagnosis algorithm identifies the fault status of the system's nodes, under the assumption that at most t of these nodes can permanently fail simultaneously. Since the introduction of the comparison model, significant progress has been made in both theory and practice associated with the original model and its offshoots. Nevertheless, the problem of efficiently identifying the set of faulty nodes when not all the comparison outcomes are available to the fault identification algorithm prior to initiating the diagnosis phase, i.e., partial syndromes, remains an outstanding research issue. In this paper, we first show how game theory can be adapted to solve the fault diagnosis problem by maximizing the payoffs of all players (nodes). We then demonstrate, using results from a thorough simulation, the effectiveness of this approach in solving the fault identification problem using partial syndromes from randomly generated diagnosable systems of different sizes and under various fault scenarios.
The reference company for this paper - a multination organization, Italian branch, that works in the domain of safety-critical systems - evaluated the major tools, which the market provides for testing safety-critical...
详细信息
ISBN:
(纸本)9728865694
The reference company for this paper - a multination organization, Italian branch, that works in the domain of safety-critical systems - evaluated the major tools, which the market provides for testing safety-critical software, as not sufficiently featured for her quality improvement goals. Consequently, in order to investigate the space of possible solutions, if any, the company's Research Lab. started an academic cooperation, which leaded to share knowledge and eventually to establish a common research team. Once we had transformed those goals in detailed technical requirements, and evaluated that it was possible to realize them conveniently in a tool, we passed to analyze, construct, and eventually utilize in field the prototype "Software Test Framework". This tool allows non-intrusive measurements on different hard-soft targets of a distributed system running under one or more Unix standard OS, e.g. LynxOS, AIX, Solaris, and Linux. The tool acquires and graphically displays the real-time flow of data, so enabling users to verify and validate software products, diagnose and resolve emerging performance problems quickly, and enact regression testing. This paper reports on the characteristics of Software Test Framework, its architecture, and results from a case study. Based on comparison of results with previous tools, we can say that Software Test Framework is leading to a new concept of tool for the domain of safety-critical software.
The goal of this paper is to characterize software technologies to test hard real-time software by focusing on measurement of CPU and memory loads, performance monitoring of processes and their threads, intrusiveness,...
详细信息
ISBN:
(纸本)9728865694
The goal of this paper is to characterize software technologies to test hard real-time software by focusing on measurement of CPU and memory loads, performance monitoring of processes and their threads, intrusiveness, and some other key features and capabilities, in the context of the Italian branch of a multinational organization, which works in the domain of safety-critical systems, from the points of view of the project managers of such an organization, on one side, and the applied researcher, on the other side. The paper first sketches on the state of the art in the field of testing technologies for safety-critical systems, then presents a characterization model, which is based on goals of the reference company, and then applies that model to major testing tools available.
The reference company for this paper - a multination organization, Italian branch, which works in the domain of safety-critical systems - evaluated the major tools that the market provides for testing safety-critical ...
详细信息
ISBN:
(纸本)9783540498605
The reference company for this paper - a multination organization, Italian branch, which works in the domain of safety-critical systems - evaluated the major tools that the market provides for testing safety-critical software, as not sufficiently featured for her quality improvement goals. Once that we had transformed those goals in detailed technical requirements, and evaluated that it was possible to realize them conveniently in a tool, we passed to analyze, construct, and eventually utilize in field the prototype "Software Test Framework". This tool allows non-intrusive parallel measurements on different hard-soft targets of a distributed system running under one or more Unix standard OS. This paper reports on the characteristics of Software Test Framework, its architecture, and results from a case study. Based on comparison of results with previous tools, we can say that Software Test Framework is leading to a new concept of tool for the domain of safety critical software.
In this paper, we contribute in solving the system-level fault diagnosis (SLD) problem under the symmetric invalidation rules. The SLD aims at determining faulty nodes based on an input syndrome that is generated comp...
详细信息
ISBN:
(纸本)9781538609583
In this paper, we contribute in solving the system-level fault diagnosis (SLD) problem under the symmetric invalidation rules. The SLD aims at determining faulty nodes based on an input syndrome that is generated comparing the outputs from the different nodes. The symmetric comparison diagnosis model assumes that tasks are assigned to pairs of nodes and the results of executing these tasks are compared. Fault nodes are identified based on the agreements and disagreements among the nodes' outputs. A limit t is assumed on the maximum number of nodes that can fail simultaneously. Various diagnosis algorithms have been developed in the last decades, but the problem of efficiently identifying the set of faulty units of a diagnosable system remained an outstanding research issue. This work introduces a new diagnosis approach for the symmetric-based comparison diagnosis problem using game theory. In this approach, faulty nodes are identified by maximizing the payoffs of all players (nodes). We have tested the new diagnosis approach using randomly generated diagnosable systems under even extreme faulty situations in which the limit t was set close to the number of nodes. The extensive simulations we have conducted proved the efficiency of the new game-theory-based diagnosis algorithm as it succeeded in guessing the fault status all nodes in all faulty situations.
Most real-world networks (including financial, health, transportation, social, citation, and sensor networks) evolve over time. Their evolution can be modeled as a series of graph snapshots that represent those networ...
详细信息
ISBN:
(纸本)9781467397957
Most real-world networks (including financial, health, transportation, social, citation, and sensor networks) evolve over time. Their evolution can be modeled as a series of graph snapshots that represent those networks at different points in time. Our distributed dynamic graph database, G*, provided efficient cluster-based storage and querying of graph snapshots by taking advantage of their commonalities. With the modern Internet of Things, however, the environment of continuously generated graph snapshots imbues classic challenges of data distribution with renewed significance. In response to this, we have extended G* to address these new challenges. In this paper we examine two snapshot placement schemes and their effects on query performance in a cloud / data center environment.
This paper describes a lightweight yet powerful approach for writing distributed applications using shared variables. Our approach, called SHAREHOLDER, is inspired by the flexible and intuitive model of information ac...
详细信息
This paper describes a lightweight yet powerful approach for writing distributed applications using shared variables. Our approach, called SHAREHOLDER, is inspired by the flexible and intuitive model of information access common to the World Wide Web. The distributed applications targeted by our approach all share a weak consistency model and loose transaction semantics, similar to a user's model of accessing email, bulletin boards, chat rooms, etc. on the Internet. The SHAREHOLDER infrastructure has several advantages. Its highly object-oriented view of shared variables simplifies their initialization and configuration. A shared variable's distribution mechanism is specified through an associated configuration object, and the programmer does not need to write any extra code to implement the sharing mechanism. These configuration objects can be initialized at run-time, allowing tremendous flexibility in dynamic control of distribution of shared variables. Finally, the programmer can treat shared variables and local variables interchangeably, thus simplifying conversion of a serial application into a distributed application.
Queueing networks with Fork-Join primitives are commonly used to model several systems with parallel activities. Unfortunately, these models are rather hard to solve analytically, as the limited results that appear in...
详细信息
Queueing networks with Fork-Join primitives are commonly used to model several systems with parallel activities. Unfortunately, these models are rather hard to solve analytically, as the limited results that appear in literature testify. The main concern of this paper is to show the equivalence between Fork-Join models, with n parallel servers and n-stage queueing networks with suitable state-dependent service rates. The equivalence relation is demonstrated by observing the isomorphism between the state transition diagrams of the Markovian processes relating to the two systems (both in the open and the short-circuited cases). This equivalence allows us to analytically solve the Fork-Join system with two and three parallel servers by analysis of the n-stage queueing network. The accuracy of this approach is verified by comparison of the approximate values with the results of a large number of simulation runs.
暂无评论