Performance and Reliability Analysis of Computer Systems

Performance and Reliability Analysis of Computer Systems: An Example-Based Approach Using the SHARPE Software Package provides a variety of probabilistic, discrete-state models used to assess the reliability and performance of computer and communication systems. The models included are combinatorial reliability models (reliability block diagrams, fault trees and reliability graphs), directed, acyclic task precedence graphs, Markov and semi-Markov models (including Markov reward models), product-form queueing networks and generalized stochastic Petri nets. A practical approach to system modeling is followed; all of the examples described are solved and analyzed using the SHARPE tool. In structuring the book, the authors have been careful to provide the reader with a methodological approach to analytical modeling techniques. These techniques are not seen as alternatives but rather as an integral part of a single process of assessment which, by hierarchically combining results from different kinds of models, makes it possible to use state-space methods for those parts of a system that require them and non-state-space methods for the more well-behaved parts of the system. The SHARPE (Symbolic Hierarchical Automated Reliability and Performance Evaluator) package is the `toolchest' that allows the authors to specify stochastic models easily and solve them quickly, adopting model hierarchies and very efficient solution techniques. All the models described in the book are specified and solved using the SHARPE language; its syntax is described and the source code of almost all the examples discussed is provided. Audience: Suitable for use in advanced level courses covering reliability and performance of computer and communications systems and by researchers and practicing engineers whose work involves modeling of system performance and reliability.

Computation of bounds for transient measures of large rewarded Markov models using regenerative randomization

Time analysis is a common approach for testing and detecting methods for the performance analysis of computer systems. In the article it is shown, that measuring and identifying performances based on a benchmark is not sucient for the proper analysis of the computer systems behavior. The response time of the process is often composed of the execution of many subprocesses or many paths of execution. Under this assumption, it is presented, that both convolution and deconvolution methods can be helpful in obtaining time distributions and modeling of complex processes. In such a modeling the analysis of measurement errors is very important and was taken into consideration. The example of using the methods in buering process is also discussed.

The First International Workshop on Teaching Performance Analysis of Computer Systems (TeaPACS1) in 2021 came to the conclusion among participants that there is a need to continue and regularly update the discussion on teaching strategies, changing the curriculum, reaching out to the systems community, facing the new generation and their new learning habits, etc.

Mean Value Analysis (MVA) has long been a standard approach for performance analysis of computer systems. While the exact load-dependent MVA algorithm is an efficient technique for computer system performance modeling, it fails to address several features of multi-core platforms. In addition, the load-dependent MVA algorithm suffers from numerical difficulties under heavy load conditions. The goal of our paper is to find an efficient and robust method which is easy to use in practice and also achieves accuracy for performance prediction for multi-core platforms. Our contributions are: We present a flow-equivalent performance model designed specifically to address multi-core computer systems. We identify the influence on the CPU demand of the effects of Dynamic Frequency Scaling (DFS) and Hyper-Threading Technology (HTT). We adopt an approximation technique to estimate resource demands to parameterize the MVA algorithm. We use a modified Conditional MVA (CMVA) algorithm to address the potential numerical instability. To validate the application of our method, we investigate a case study of an e-commerce web server which is equipped with diverse classes of user requests. We show that our method achieves better accuracy compared with other commonly used MVA algorithms.

This paper assesses the state of the art in cost-benefit analyses of computer graphics systems and suggests an approach for developing improved methodology. Cost-benefit analyses are distinguished from analyses of system performance in that the latter is directed at optimizing system performance at a given level of investment, while the former is directed at justifying the investment itself.Computer graphic system design alternatives are first outlined. Then methods of analyzing the performance and costs of computer systems in general and graphic systems in particular are discussed. With this information it is shown how cost-effectiveness analyses may be performed. The next crucial step is to conduct benefit analysis, an ill-defined art. The results of benefit analysis must be combined with cost-effectiveness analysis in order to perform the desired cost-benefit analysis.An experimental methodology is suggested for better performing benefit analyses of computer graphics systems. A more rigorous formulation of the cost-benefit procedure is then outlined. No attempt is made in this report to actually perform such an analysis.

Performance modeling and analysis of computer systems and networks

Trace-driven simulation is an important aid in performance analysis of computer systems. Capturing address traces for these simulations is a difficult problem for single processors and particularly for multicomputers. Even when existing trace methods can be used on multicomputers, the amount of collected data typically grows with the number of processors, so I/O and trace storage costs increase. A new technique is presented in this paper which modifies the executable code to dynamically collect the address trace from the user code and analyzes this trace during the execution of the program. This method helps resolve the I/O and storage problems and facilitates parallel analysis of the address trace. If a trace stored on disk is desired, the generated trace information can also be written to files during execution, with a resultant drop in program execution speed. An initial implementation on the Intel iPSC/2 hypercube multicomputer is detailed, and sample simulation results are presented. The effect of this trace collection method on execution time is illustrated.

A new “hybrid” analytic framework, based on the principle of maximum entropy, is used to derive a closed form expression for the queue length distribution of a G/G/1 finite capacity queue. It is shown that “Birth-Death” homogeneous recursions for a single resource queue are special cases of maximum entropy “one-step” transitions which can be applied either in an operational or stochastic context. Furthermore, an “equivalence” relationship is used to analyse two-stage cyclic queueing networks with general service times, and favourable comparisons are made with global balance and approximate results. Numerical examples provide useful information on how critically system behaviour is affected by the distributional form of interarrival and service patterns. Comments on the implication of the work to the performance analysis and aggregation of computer systems are included.

The adoption of a “high level” perspective in the design of a component-based application, without considering the specific features of some underlying supporting platform, has the advantage of focusingon the relevant architectural aspects and reasoning about them in a platform independent way, omitting unnecessary details that could even not be known at the earliest development stages.On the other hand, many of the details that are typically neglected in this high-level perspective must necessarily be taken into account to obtain a meaningful evaluation of different architectural choices in terms of extra-functional quality attributes, like performance or reliability. Toward the reconciliation of these two contrasting needs, we propose a model-based approach whose goal is to support the derivation of sufficiently detailed prediction models from high level models of component-based systems, focusing on the prediction of performance and reliability. We exploit for this purpose a refinement mechanism based on the use of model transformation techniques.

Performance analysis of pressurized irrigation systems operating on-demand using flow-driven simulation models

The paper is concerned with the performance analysis of multiple-access DS unslotted ALOHA systems with fix packet length in synchronous (chip slotted) channels with single spreading code or CDMA coding. In contrast with previous work the present authors focus on the performance analysis of a system that can have given auto- and crosscorrelation sidelobes between the spreading codes. They present a novel approach based on the multivariable probability generator function of the level of interferences which facilitates the calculation of packet error probability for different systems without error correcting codes. In the paper the lower bound of the throughput is calculated. And its asymptotic behaviour is considered, when the length of the DS code goes to infinity. Generally the chip spread ALOHA model is analyzed, but the results can be used in the case of bit spread ALOHA systems. >

PurposeDegraded failures and sudden critical failures are quite prevalent in industries. Degradation processes commonly belong to Weibull family and critical failures are found to follow exponential distribution. Therefore, it becomes important to carry out reliability and availability analysis of such systems. From the reported literature, it is learnt that models are available for the situations where the degraded failures as well as critical failures follow exponential distribution. The purpose of this paper is to present models suitable for reliability and availability analysis of systems where the degradation process follows Weibull distribution and critical failures follow exponential distribution.Design/methodology/approachThe research uses Semi-Markov modeling using the approach of method of stages which is suitable when the failure processes follow Weibull distribution. The paper considers various states of the system and uses state transition diagram to present the transition of the system among good state, degraded state and failed state. Method of stages is used to convert the semi-Markov model to Markov model. The number of stages calculated in Method of stages is usually not an integer value which needs to be round off. Method of stages thus suffers from the rounding off error. A unique approach is proposed to arrive at failure rates to reduce the error in method of stages. Periodic inspection and repairs of systems are commonly followed in industries to take care of system degradation. This paper presents models to carry out reliability and availability analysis of the systems including the case where degraded failures can be arrested by appropriate inspection and repair.FindingsThe proposed method for estimating the degraded failure rate can be used to reduce the error in method of stages. The models and the methodology are suitable for reliability and availability analysis of systems involving degradation which is very common in systems involving moving parts. These models are very suitable in accurately estimating the system reliability and availability which is very important in industry. The models conveniently cover the cases of degraded systems for which the model proposed by Hokstad and Frovig is not suitable.Research limitations/implicationsThe models developed consider the systems where the repair phenomenon follows exponential and the failure mechanism follows Weibull with shape parameter greater than 1.Practical implicationsThese models can be suitably used to deal with reliability and availability analysis of systems where the degradation process is non-exponential. Thus, the models can be practically used to meet the industrial requirement of accurately estimating the reliability and availability of degradable systems.Originality/valueA unique approach is presented in this paper for estimating degraded failure rate in the method of stages which reduces the rounding error. The models presented for reliability and availability analyses can deal with degradable systems where the degradation process follows Weibull distribution, which is not possible with the model presented by Hokstad and Frovig.

Performance and stability analysis of discontinuous PWA systems by piecing together PWQ functions

This paper proposes a new goal-oriented (GO) method for reliability analysis of repairable systems with multiple-input and multi-function component (MIMFC). First, a new GO operator combination, which is composed of a new function GO operator and a new auxiliary GO operator, is created to represent MIMFC. The new function GO operator named as Type 22 operator is created to represent MIMFC itself, and the auxiliary GO operator named as Type 15B operator is created to represent multi-condition control signals of MIMFC. Then, GO operation formulas of the new GO operator combination are deduced based on logical relationships among inputs, outputs, and the component itself. The reliability analysis process of the new GO method is formulated. Furthermore, this new GO method is applied for the first time in steady availability analysis and qualitative analysis of the fan drive system of a power-shift steering transmission. Finally, the results obtained by the new GO method are compared with the results of fault tree analysis (FTA) and Monte Carlo simulation (MCS), and the comparison results show that this new GO method is reasonable and advantageous in reliability analysis of repairable systems with MIMFC. Moreover, the analysis process shows that it is more advantageous in the aspect of building system models and conducting reliability analysis. Overall, this paper not only improves the basic theory of the GO method and expands the application of the GO method, but it also provides a new approach for reliability analysis of repairable systems with MIMFC.