Olivia Das

Enhanced Modeling and Solution of Layered Queueing Networks

Layered queues are a canonical form of extended queueing network for systems with nested multiple resource possession, in which successive depths of nesting define the layers. The model has been applied to most modern distributed systems, which use different kinds of client-server and master-slave relationships, and scales up well. The layered queueing network (LQN) model is described here in a unified fashion, including its many more extensions to match the semantics of sophisticated practical distributed and parallel systems. These include efficient representation of replicated services, parallel and quorum execution, and dependability analysis under failure and reconfiguration. The full LQN model is defined here and its solver is described. A substantial case study to an air traffic control system shows errors (compared to simulation) of a few percent. The LQN model is compared to other models and solutions, and is shown to cover all their features.

Throughput Analysis of Opportunistic Access Strategies in Hybrid Underlay—Overlay Cognitive Radio Networks

In cognitive radio networks, it is important to effectively use the under-utilized spectrum resources without affecting the primary users. In an underlay system, secondary users are allowed to share the channel simultaneously with primary users (with the restriction on interference level) but not in an overlay system. In this article, we consider a system where a secondary user can switch between overlay and underlay modes of operation in order to improve its throughput with limited sensing capability (i.e. sensing only one channel at a time). The results based on Markov chain analysis are satisfactorily verified using Monte-Carlo simulation. It is found that proper selection of transmission mode can provide greater improvement in throughput for a secondary user. The mode selection depends on the transition characteristics of primary users and the throughput ratio between the two modes of operation.

Cooperative subcarrier and power allocation for a two-hop decode-and-forward OFCMD based relay network

In this article, subcarrier and power allocation schemes are proposed and analyzed for different scenarios for a two-hop decode-and-forward OFCDM based relay network. In subcarrier allocation, the effect of considering the channel state information (CSI) of source-base station and source-relay link are evaluated in a cooperative diversity system. Results show that allocation of subcarriers based on source-relay node CSI provides better BER performance at higher Eb/No, and at lower Eb/No, both the source-relay and source-base station links need to be considered. From our numerical simulation, we also noticed that the cross-over Eb/No, point (around which frequency spreading gives better performance than time spreading) moves towards the lower Eb/No, when the subcarrier allocation is done giving more weight to source-base station link rather than the source-relay link which provides additional flexibility in operating environment for OFCDM systems. In power allocation, a cooperative power allocation ratio lambda (=source node power/total power) is defined and BER performance is evaluated for different values of lambda in the relay network. It is found that there exists an optimal power allocation ratio for different operating environment such as source-to-relay channel gains and time-frequency spreading factors. It is reported that: (a) when all three channels (source-to-relay, source-to-destination and relay-to-destination) have equal gains, power ratio is found to be lambda ap 0.8 (i.e., 80% and 20% of the total power is distributed among source and relay node respectively). The performance degrades at much faster rate when lambda increases above the optimal value at higher Eb/No. On the other hand, the performance remains almost the same when the decrement in lambda is less than the optimal value. (b) For a network with stronger source-to-relay link, the optimal lambda remains almost the same as the case with equal channel gains at higher Eb/No; however, the optimal power ratio moves toward lower value of lambda of 0.65 at lower b/No. (c) The optimal lambda remains almost the same with different time-frequency spreading factors.

The fault-tolerant layered queueing network model for performability of distributed systems

Proliferation of large and complex fault-tolerant distributed systems in recent years has stimulated the combined modelling of performance and dependability of such systems. For large systems it may be very expensive to compute valid performance estimates to be used in the combined performability measures. This paper considers a class of systems with a fault-tolerant client-server structure, in which efficient layered queueing network models can be combined with AND-OR graph analysis. The model solution can be obtained without solving very large Markovian models.

Evaluating layered distributed software systems with fault-tolerant features

Abstract Proliferation of large and complex fault-tolerant distributed systems in recent years has stimulated the combined modelling of performance and dependability of such systems. For large systems it may be very expensive to compute valid performance estimates to be used in the combined performability measures. This work considers two different classes of fault-tolerant client–server systems, in which efficient layered queueing network models can be combined with AND–OR graph analysis. The fault-tolerance in the first class is achieved by incorporating stand-by servers and a mechanism to re-direct the service requests at the time of failure whereas fault-tolerance in the second class is achieved by incorporating redundant servers that can all be used at the same time with the service load being distributed equally among them. The model solutions in both the cases can be obtained without solving very large Markovian models.

Opportunistic Channel Sharing Based on Primary User Transition Probabilities in Dual Mode Cognitive Radio Networks

In cognitive radio networks, it is important to effectively use the under-utilized spectrum resources without affecting the primary users. In an underlay system, cognitive users are allowed to share the channel simultaneously with primary users with the restriction on interference level but not in an overlay system. In this paper, we consider a system where cognitive users can switch between overlay and underlay modes of operation in order to improve their throughput. The results, based on Markov chain analysis, are satisfactorily verified using Monte-Carlo simulation. It is shown that proper selection of transmission mode can provide greater improvement in throughput for cognitive users. If a primary user occupies the channel for a longer (shorter) period, then the system should allow the cognitive users to choose underlay (overlay) mode for throughput advantage. The exact strategy (or mode) switching thresholds can be found from our analysis and it depends primarily on the transition probabilities of the primary users and throughput ratio during underlay/overlay transmission.

Modeling the coverage and effectiveness of fault-management architectures in layered distributed systems

Increasingly, fault-tolerant distributed software applications use a separate architecture for failure detection instead of coding the mechanisms inside the application itself. Such a structure removes the intricacies of the failure detection mechanisms from the application, and avoids repeating them in every program. However, successful system reconfiguration now depends on the management architecture (which does both fault detection and reconfiguration), and on management subsystem failures, as well as on the application. This paper presents an approach which computes the architecture-based system reconfiguration coverage simultaneously with its performability.

Web Application Performance Modeling Using Layered Queueing Networks

In this paper, a Layered Queueing Network (LQN) performance model is used for studying an Apache-PHP web application with PostgreSQL backend-database. Performance evaluation is done by obtaining load test measurements and by solving the LQN model. Model validation is performed by comparing the model results with the load test results. With average error of 3.77% for throughput and 12.15% for response times the model is shown to capture the web application@?s performance. Furthermore, performance analysis is done to determine the system configuration which would ease the identified bottleneck resource.

Dependable LQNS: a performability modeling tool for layered systems

Dependable LQNS is a software tool for modeling and evaluating performability of fault-tolerant layered distributed applications that use a separate architecture for failure detection and reconfiguration. It takes into account the effects of management architecture, application software architecture, failure of management and application components in the dependability computation. It uses a combination of minpath algorithms, AND-OR graphs, noncoherent fault trees and Layered Queueing modeling in the analysis.

An Availability Model of a Virtual TMR System with Applications in Cloud/Cluster Computing

Three important factors in dependable computing are cost, error correction and high availability. In this paper we will focus on assessing a proposed model that encapsulates all three important factors and a virtual architecture that can be implemented in the IaaS layer of cloud computing. The proposed model will be assessed against a popular existing architecture (Triple Modular Redundant System TMR) and the availability analysis done with Fault-Trees combined with Markov Chains. These experiments will demonstrate that the virtualization of the TMR system using the architecture that we have proposed, will achieve almost the same level of availability/reliability and cost, along with the inherent advantages of virtual systems. Advantages include faster system restart, efficient use of resources and migration.