Veena B. Mendiratta

Characterizing session initiation protocol (SIP) network performance and reliability

The Session Initiation Protocol (SIP) has emerged as the preferred Internet telephony signaling protocol for communications networks. In this capacity, it becomes increasingly essential to characterize both the performance and the reliability of the signaling entities utilizing the protocol. We provide an analytical look at the performance of a SIP network as well as a reliability model of SIP servers.

Survivability analysis of telephone access network

The telecommunications industry has achieved high reliability and availability for telephone service over decades of development. However, the current design does not aim at providing service survivability when a local switching office fails due to catastrophic damage. In this paper, several survivable architectures for telephone subscriber network are proposed based on common survivability principles. In order to quantitatively assess the effectiveness of design alternatives, a set of analytical models are developed to derive various survivability measures. Numerical results are provided to show how a comprehensive understanding of the system behavior after failure can be achieved through different survivability aspects.

Reliability analysis of clustered computing systems

Clustered computing systems, using commercially available computers networked in a loosely-coupled fashion, can provide high levels of reliability if appropriate levels of error detection and recovery software are implemented in the middleware and application layers. In this paper, we present a modeling approach for analyzing the hardware and software reliability of clustered computing systems. The clustered system is modeled as an irreducible Markov chain with working and failed states, and intermediate recovery states. The failure and recovery behavior is characterized in terms of the frequency and duration of fault recoveries and outages for a single processor in the cluster and for the entire clustered system. We apply the model to a telecommunication switching system application that uses the Lucent Technologies Reliable Clustered Computing product. The model results are presented for a range of values of the processor failure rate and the fault recovery coverage factor.

Mission critical communication networks for railways

Communication networks in the railway sector are critical to the operation of the system and have stringent requirements for reliability and safety. These types of networks are commonly characterized as “mission critical.” Further, rail communication networks have requirements for interoperability with legacy technology and long life cycle support. Many of the European railways operate trackside Global System for Mobile Communications-Railway (GSM-R) wireless networks; GSM-R is based on the GSM standard with railway-specific features. The railways have started to look at Long Term Evolution (LTE) as a potential future replacement system for GSM-R. This paper presents the role of communication networks in railway operations, the resulting unique requirements for mission critical rail networks, and current trends in railway telecommunications. A brief tutorial on GSM-R is provided. We then present the LTE network architecture and assess the suitability of LTE to meet the requirements of the railway sector, with a special focus on reliability. The paper focuses primarily on mainline rail networks; however, much of what is presented also is applicable to urban rail networks.

The Importance of Outlier Relationships in Mobile Call Graphs

Mobile phones have become one of the primary tools for individuals to communicate, to access data networks, and to share information. Service providers collect data about the calls placed on their network, and these calls exhibit a large degree of variability. Providers model the structure of the relationships between network subscribers as a mobile call graph. In this paper, we apply a new measure to quantify by how much a relationship between users in a mobile call graph deviate from an average relationship. This measure is used to explore the connection between calling behaviors and the complex structure mobile call graphs take. We study a large call graph from a major service provider and learn that distant, outlier relationships play the largest role in maintaining connectivity between cellular users, and that calling features of users more strongly influence tie variation compared to social features. We also observe a rapid decay of its massively connected component as outlier ties are removed.

Assessing the reliability impacts of software fault-tolerance mechanisms

Telecommunications systems are characterized by highly stringent reliability requirements for system availability and defect rate. A combination of approaches is used to achieve high software reliability, namely, fault avoidance, fault removal and implementation of fault-tolerant mechanisms. This paper focuses on the implementation of software fault-tolerant mechanisms and analyzes the impact of these mechanisms on software reliability. Based on field data on the frequency of invocation of some fault-tolerant mechanisms, we present an escalating recovery model for predicting the impact of these mechanisms on lost calls. The key parameters of the model are: the software fault recovery coverage factor; the proportion of successful recoveries at each level and the calls lost at each recovery level. The output of the model is a distribution and average of the number of lost calls per software error. The applicability of this model to systems with high reliability has been validated; the applicability of the model to less reliable systems is an area for future work.

Cloud Incident Data: An Empirical Analysis

This paper presents an empirical analysis of cloud incidents reported in the Cloutage.org database. The trend, causes, and impact of three types of incidents, namely, Outage, Vulnerability, and, failure during automatic updates (Auto Fail) were examined. Service availability was also analyzed based on the outage duration data. The analysis suggested that: (i) Outages and Vulnerabilities grow exponentially, while Auto Fail incidents show only a linear increase, (ii) Outages are caused by various sources of failures, Vulnerabilities primarily due to the lack of filtering inputs and most Auto Fail incidents are false positives, (iii) Many outages affected multiple, related services, some cascaded into additional ones during resolution, and the impact of some transcended organizational boundaries, and (iv) Availability of cloud services is less than 99%, and for free services such as Email it could be as low as 84%. The paper concludes with a summary of key observations and offers recommendations along three dimensions to avoid and alleviate the impact of cloud incidents.

Architectures and disaster recovery strategies for survivable telecommunications services

The events of September 11, 2001, not only changed the public worldview of security and trust but also the telecommunications industrys view of how it must address Homeland Security requirements for service survivability and recovery. Historically, public telecommunications service has evolved to be highly reliable through hardware and software redundancy, fault recognition, and containment; it now needs to be highly survivable to limit the extent of service loss after a catastrophic loss of infrastructure. This paper defines service survivability and reliability and highlights the similarities and differences in procedures and mechanisms that are required for each. The topic is discussed from two perspectives: (1) architectural principles that support the minimization of service loss and (2) disaster recovery plans and procedures that support rapid, predictable restoration of service after a loss. Alternative architectures with improved survivability are examined within the context of the current circuit-based public switched telephone network (PSTN) and other technologies, such as wireless and voice over Internet protocol (VoIP).

Rejuvenation with workload migration

A five-state model of software rejuvenation is introduced that divides the working state into three sub-states: working, vulnerable, and preparing. The preparing state models the period during which workload is drained from software elements that are about to be rejuvenated. We compared our model to a four-state model that only has working and vulnerable sub-states.

Reliability of IMS architecture

The IP multimedia subsystem (IMS) enables an access-agnostic next generation network (NGN) that integrates wireless, wireline and Internet technologies. IMS provides for real-time voice and data services over a packet-switched domain via a session initiation protocol (SIP) infrastructure. It also offers a cost-effective platform for the creation and deployment of IP-based multimedia services. Given the wide scope of services that can be offered with IMS it is important to develop a framework to analyze the reliability of these services. Using an IMS reference architecture, this paper presents a network level failover analysis using the IMS architecture capabilities and shows its impact on improving service reliability. We show how to analyze the end-to-end service reliability in terms of detection and recovery mechanisms and redundancy.