Using Fault Injection and Modeling to Evaluate the Performability of Cluster-Based Services
نویسندگان
چکیده
We propose a two-phase methodology for quantifying the performability (performance and availability) of cluster-based Internet services. In the first phase, evaluators use a fault-injection infrastructure to measure the impact of faults on the server’s performance. In the second phase, evaluators use an analytical model to combine an expected fault load with measurements from the first phase to assess the server’s performability. Using this model, evaluators can study the server’s sensitivity to different design decisions, fault rates, and environmental factors. To demonstrate our methodology, we study the performability of 4 versions of the PRESS Web server against 5 classes of faults, quantifying the effects of different design decisions on performance and availability. Finally, to further show the utility of our model, we also quantify the impact of two hypothetical changes, reduced human operator response time and the use of RAIDs.
منابع مشابه
Using Fault Injection to Evaluate the Performability of Cluster-Based Services
We propose a two-phase methodology for quantifying the performability (performance + availability) of cluster-based Internet services. In the first phase, evaluators use a fault-injection infrastructure to measure the impact of faults on the server’s performance. In the second phase, evaluators use an analytical model to combine an expected fault load with measurements from the first phase to a...
متن کاملEvaluating the Impact of Communication Architecture on the Performability of Cluster-Based Services
We consider the impact of different communication architectures on the performability (performance + availability) of cluster-based servers. In particular, we use a combination of fault-injection experiments and analytic modeling to evaluate the performability of two popular communication protocols, TCP and VIA, as the intra-cluster communication substrate of a sophisticated Web server. Our ana...
متن کاملPerformability Evaluation of Networked Storage Systems Using N-SPEK
This paper introduces a new benchmark tool for evaluating performance and availability (performability) of networked storage systems, specifically storage area network (SAN) that is intended for providing block-level data storage with high performance and availability. The new benchmark tool, named N-SPEK (Networked-Storage Performability Evaluation Kernel module), consists of a controller, sev...
متن کاملTowards Performability Modeling of Software Rejuvenation
In this paper, we discuss issues in performability modeling of \software rejuvenation," a form of software fault tolerance based on occasionally cleaning up the operational environment. System factors which play a key role in such a model are identi ed. Among these, we comment on two issues of particular interest when modeling software rejuvenation: (1) the representation of the degradation in ...
متن کاملA Case Study: Validation of Guidance Control Software Requirements for Completeness, Consistency and Fault Tolerance
In this paper, we discuss a case study performed for validating a Natural Language (NL) based software requirements specification (SRS) in terms of completeness, consistency, and fault-tolerance. A partial verification of the Guidance and Control Software (GCS) Specification is provided as a result of analysis using three modeling formalisms. Zed was applied first to detect and remove ambiguity...
متن کامل