Rapid adoption of smartphones and tablets in recent years has changed the way people communicate with each other and operate in the work environment. The use of personal devices in the enterprise has become commonplace. These devices have given workers the flexibility of being mobile and still participate in collaboration sessions. However, transferring sessions between devices is not very seamless and causes disruptions in the sessions, resulting in an unsatisfactory user experience. This work presents an architecture and implementation for seamless session mobility across different types of devices and hardware/software platforms. This architecture is not restricted to a specific collaboration server and can accommodate multi-vendor environments.

There are two broad categories of approaches used for checkpointing: application-transparent and application-assisted. Typically, application-assisted approaches provide a more flexible and light-weight mechanism but require changes to the application. Although most applications run well under virtualization (e.g. Xen which is being adopted widely), the addition of application-assisted checkpointing - used for high availability - causes performance problems. This is due to the overhead of key system calls used by the checkpointing techniques under virtualization. To overcome this, we introduce the notion of hypervisor-assisted application checkpointing with no changes to the guest operating system. We present the design and a Xen-based implementation of our family of application checkpointing techniques. Our experiments show performance improvements of 4× to 13× in the primitives used for supporting high availability compared to purely user-level approaches.

Virtualization provides enormous economic and ecological benefits by enabling server consolidation and supporting low-cost green data centers. A key component of the hypervisor in a virtualized system is the scheduler. A typical scheduler provides parameters to influence the hypervisor's behavior. Specifically, an application's performance on the popular open-source Xen virtualization platform can be influenced by tuning its scheduler behavior using the weight, cap, and processor pinning variables. However, determining which parameters to tune and how to do that is non-trivial. In this paper, we introduce XenTune, a monitoring tool for the credit scheduler in Xen that helps in understanding application behavior in scheduler terms and assists in determining scheduler parameters. We demonstrate how the tool is used with application domains - specifically a media application domain. We demonstrate experimental results using a real work-load that shows the considerable benefits of correct scheduler parameter choices. XenTune had helped in the design of a recently proposed scheduler S and in this paper we show how scheduler S interacts with the media application to optimize its performance.

Virtualization technology enables server consolidation and has given an impetus to low-cost green data centers. However, current hypervisors do not provide adequate support for real-time applications, and this has limited the adoption of virtualization in some domains. Soft real-time applications, such as media-based ones, are impeded by components of virtualization including low-performance virtualization I/O, increased scheduling latency, and shared-cache contention. The virtual machine scheduler is central to all these issues. The goal in this paper is to adapt the virtual machine scheduler to be more soft-real-time friendly.

Log monitoring techniques to characterize system and user behavior have gained significant popularity. Some common applications of study of systems logs are syslog mining to detect and predict system failure behavior, web log mining to characterize web usage patterns, and error/debug log analysis for detecting anomalies. In this paper, we discuss our experiences with applying log mining techniques to characterize the behavior of large enterprise telephony systems. We aim to detect, and in some cases, predict system anomalies. We describe the problems encountered in the study of such logs and propose some solutions. The key differentiator of our solutions is the use of individual message frequencies to characterize system behavior and the ability to incorporate domain-specific knowledge through user feedback. The techniques that we propose are general enough to be applicable to other systems logs and can easily be packaged into automated tools for log analysis.

Multi-hop ad-hoc wireless networks need additional software at each node to provide routing functionality. There is significant benefit in supporting legacy devices (i.e., devices that do not have any additional routing software) in such networks. In this paper, we present a system and techniques for legacy device support in multi-hop ad-hoc wireless networks. One conceptual approach is to enable access point-like support at each multi-hop ad-hoc node that allows the legacy devices to associate in infrastructure mode. In contrast, we present a novel scalable architecture that extends the benefits of path redundancy inherent in multi-hop ad-hoc networks to legacy devices and also supports mobility. Using the legacy device as a message reflector in a transparent manner allows the system to make decisions in a light-weight distributed fashion. We describe our architecture, protocols and a prototype system. We also present some analysis and observations from the prototype.

In this paper, the authors motivate the interesting characteristics of small-to-medium enterprises and study the problem of wireless IP telephony for such deployments. While small enterprises are characterized by few network nodes and small deployment areas, they also demand good IP telephony quality, ease of deployment, and low management and configuration overhead. The authors evaluate the feasibility of peer-to-peer telephony over a multi-hop wireless ad-hoc network for such small enterprises. The authors designed a system and built a prototype using off-the-shelf components. In this paper, the authors describe the implementation and deployment of this system. Several fundamental routing protocol issues needed to be resolved to develop a stable system and the authors present them here. The authors also report the results of a simulation study they conducted to establish the configuration and performance parameters for such systems, and identify the limits to the use of multi-hop ad-hoc wireless for peer-to-peer IP telephony for small enterprises.

Object Persistence is an important feature of Object-oriented languages. The C++ language specification does not include or discuss any method of providing persistence for C++ objects. Several schemes have been developed for adding persistence to C++. Some of them require persistent objects to be allocated and treated differently than non-persistent objects, while some others require the programmer to provide vital parts of the persistence mechanism. It is desirable to make the persistence feature transparent, but the nature of C++ makes it difficult. This paper discusses in detail the various interesting language issues to be considered for adding persistence to C++ and how they lead to the design of the reflective object-checkpointing library, MemberAnalyzer.

We propose a mechanism, based on fault tolerant request IDs piggybacked on GIOP messages, that can be used to enable fault tolerant CORBA invocations spanning multiple heterogeneous ORBs. The proposed mechanism supports replication at both server and client sides.

We study the performance of a unified power control and forward error correction control algorithm for mobile radio systems that attempts to maximize system capacity and minimize transmitter power utilization while satisfying user defined quality of service constraints. The algorithm is distributed, with each individual transmitter-receiver pair determining its own operating point with respect to power and error control parameters. Simulation results from a cellular system with mobile users show that this unified algorithm outperforms the schemes that use power control alone.

Summary form only given, as follows. A large number of tools and techniques have been developed in the past to achieve a 24/spl times/7 system availability (24 hours a day and 7 days a week) by reducing unscheduled system downtime due to failures. However, a highly available or fault-tolerant system may still have to be taken off-line for software and hardware updates, maintenance and rejuvenation. Therefore, the scheduled downtime for maintenance could become the major source of system unavailability. One big challenge in a highly available system is to keep the system running while it is undergoing software updates or bug fixes. In this paper, we describe a tool that can be used to perform an online update of software in a cluster environment. The tool consists of a protocol compiler (stgen) and a library (libst) for marshaling and unmarshaling data during a software update. The tool has the ability to transfer complex data structures between two processes even if the data definitions in the two processes are different. The data transfer format is machine-independent. Hence, the tool can transfer data across processes running on different machine types. The paper describes some real-life applications of the tool and presents performance measurements of the tool for these applications.

Describes the design and analysis of RobustWeb, a scalable and fault-tolerant World Wide Web server cluster that is based on HTTP redirection. The system consists of a set of N back-end document servers and one or more redirection servers which receive the HTTP requests and redirect them to the document servers. A load distribution algorithm is used for initial distribution of the documents on the servers. Given a specific degree of replication k, the distribution algorithm guarantees that at least k replicas of each document are present after document distribution is complete. The redirection servers redirect requests to one of the replicas with a pre-computed redirection probability. When a server fails, the redirection probabilities are recomputed using a novel algorithm based on network flow. Theis enables the load to be approximately balanced among the remaining servers, allowing for graceful degradation of the service in the event of failures. A preliminary prototype of RobustWeb has been implemented.

Reliability of compute-intensive applications can be improved by introducing fault tolerance into the system. Algorithm based fault tolerance (ABFT) is a low-cost scheme which provides the required fault tolerance to the system through system level encoding. In this paper, we propose randomized construction techniques, under an extended model, for the design of ABFT systems with the required fault tolerance capability. The model considers failures in the processors performing the checking operations.

Algorithm-based fault tolerance (ABFT) is a technique which improves the reliability of a multiprocessor system by providing concurrent error detection and fault location capability to it. It encodes data at the system level and modifies the algorithm to operate on the encoded data in order to expose both transient and permanent faults in any processor. Work done till now in this area takes care of only the fault detection and location part of the problem. However, if spare processors are not available, then after a faulty processor has been located, the work initially assigned to it has to be mapped to some nonfaulty processors in the system in such a way that the fault tolerance capability of the system is still maintained with as small a degradation in performance as possible. In this paper, we propose an integrated deterministic solution to the above problem which combines concurrent error detection and fault location with graceful degradation. There exists no previous deterministic ABFT method for the design of general t-fault locating systems, even for the case of t=1. We propose a general method for designing one-fault locating/s-fault detecting systems. We use an extended model for representing ABFT systems. This model considers the processors computing the checks to be a part of the ABFT system, so that faults in the check computing processors can also be detected and located using a simple diagnosis algorithm, and the checks can be mapped to other nonfaulty processors in the system.

In the last decade the growth of wireless communication has been rapid. One critical issue in wireless systems is the efficient encoding of data to ensure correctness in the presence of noise and multipath fading. We propose some encoding schemes which span the FEC/ARQ spectrum and analyze them with respect to throughput and delay. We conclude with an adaptive algorithm for choosing the optimal encoding scheme based on observed error rates.

Algorithm-based fault tolerance (ABFT) is a technique for improving the reliability of a multiprocessor system by providing concurrent error detection and fault location capability to it. In this paper, we propose the first integrated solution to the problem of fault detection, location and graceful degradation in ABFT systems. Unlike most previous methods, we use an extended model for representing ABFT systems, which allows faults to occur in check computing processors.>

Molecular dynamics (MD) simulations play an important role in the study of conformations and dynamics of proteins and nucleic acids. This paper suggests ways of mapping the molecular dynamics calculations to various types of parallel architectures, e.g. mesh array, linear array, such that the inherent redundancy involved in the simulation algorithm can be used to make the computation fault tolerant without much overhead.>

To improve the reliability of computeintensive applications run on multiprocessor architec tures, fault tolerance is introduced into the system with on-line detection and location of faults. This can be achieved by a low-cost scheme, called Algorithm-based fault tolerance (ABFT), which encodes data at the system level and modifies the algorithm to operate on the encoded data. The resultant encoded output data is checked for correctness by some checks. In this pa per we present an extended model for representing and designing ABFT systems. The model takes into con sideration the processors evaluating the checks. We propose a design method which considers the proces sors computing the checks to be a part of the ABFT system and guarantees concurrent error detection even in the presence of faults in these processors, unlike most methods presented earlier.