Catálogo de publicaciones - libros

In 2003, Shim proposed an efficient ID-based authenticated key agreement protocol based on Weil pairings [1]. Sun et al. raised the potential of a man-in-the-middle attack in [2]. In 2004, Ryu et al. proposed an efficient ID-based authenticated key agreement protocol from pairings [3]. In 2005, however, Boyd et al. noted security problems of Ryu et al.’s protocol in [4]. In 2005, Yuan et al. also pointed out the same weakness [5] in Ryu et al.’s protocol. Then, they proposed a new protocol that combines Ryu et al.’s protocol with Shim’s protocol. In this paper, we demonstrate that Shim’s protocol does not provide KGC forward secrecy, then we propose a more efficient and secure protocol which does provide such security. As a result, our protocol does not need an additional ECC point-addition unlike Yuan et al.’s protocol and our’s can generate two secure session key to perform the secure message transmission.

Self-managing systems require continuous monitoring to ensure correct operation. Detailed monitoring is often too costly to use in production. An alternative is adaptive monitoring, whereby monitoring is kept to a minimal level while the system behaves as expected, and the monitoring level is increased if a problem is suspected. To enable such an approach, we must model the system, both at a minimal level to ensure correct operation, and at a detailed level, to diagnose faulty components. To avoid the complexity of developing an explicit model based on the system structure, we employ simple statistical techniques to identify relationships in the monitored data. These relationships are used to characterize normal operation and identify problematic areas.

We develop and evaluate a prototype for the adaptive monitoring of J2EE applications. We experiment with 29 different fault scenarios of three general types, and show that we are able to detect the presence of faults in 80% of cases, where all but one instance of non-detection is attributable to a single fault type. We are able to shortlist the faulty component in 65% of cases where anomalies are observed.

The number of processors embedded in high performance computing platforms is growing daily to solve larger and more complex problems. The logical network topologies must also support the high degree of scalability in dynamic environments. This paper presents a scalable and fault tolerant topology called binomial graph (BMG). BMG provides desirable topological properties in terms of both scalability and fault-tolerance for high performance computing such as reasonable degree, regular graph, low diameter, symmetric graph, low cost factor, low message traffic density, optimal connectivity, low fault-diameter and strongly resilient. Several fault-tolerant routing algorithms are provided on BMG for various message types. More importantly, BMG is able to deliver broadcast messages from any node within () steps.

We present a novel implementation of the () from the original hierarchy of Chandra-Toueg oracles. Previous implementations of have assumed models of partial synchrony where point-to-point message delay is bounded and/or communication is reliable. We show how to implement this important oracle under even weaker assumptions using Average Delayed/Dropped (ADD) channels. Briefly, all messages sent on an ADD channel are or . All non-privileged messages can be arbitrarily delayed or even dropped. For each run, however, there exists an unknown window size , and two upper-bounds and , where bounds the average delay of the last privileged messages, and bounds the ratio of non-privileged messages to privileged messages per window.

This paper proposes a novel analytical modeling approach to investigate the performance of five prominent adaptive routings in wormhole-switched 2-D tori fortified with an effective scheme suggested by Chalasani and Boppana [1], as an instance of a fault-tolerant method. This scheme has been widely used in the literature to achieve high adaptivity and support inter-processor communications in parallel computers due to its ability to preserve both communication performance and fault-tolerant demands in such networks. Analytical results of the model are confirmed by comparing with those obtained through simulation experiments.

In large cluster-based machines, fault-tolerance in the interconnection network is an issue of growing importance, since their increasing size rises the probability of failure. The topology used in these machines is usually a fat-tree. This paper proposes a new distributed fault-tolerant routing methodology for fat-trees. It does not require additional network hardware. It is scalable, since the required memory, switch hardware and routing delay do not depend on the network size. The methodology is based on enhancing the Interval Routing scheme with exclusion intervals. Exclusion intervals are associated to each switch output port, and represent the set of nodes that are unreachable from this port after a failure appears. We propose a mechanism to identify the exclusion intervals that must be updated after detecting a failure, and the values to write on them. Our methodology is able to support a relatively high number of network failures with a low degradation in network performance.

The rVsAll rollback-recovery protocol assures that consistency models, called session guarantees, are provided for mobile clients and unreliable servers. The protocol is optimized according to the required consistency model by taking into account properties of session guarantees and integrating their consistency management with known rollback-recovery techniques: message-logging and checkpointing. We have proved the correctness of rVsAll protocol in terms of safety and liveness properties. These properties assert that clients access object replicas maintained by servers according to the required session guarantee regardless of server’s failures, and state that each access operation issued by clients will eventually be performed (in a finite time), even in the presence of server’s failures.

In this paper, we show that the proposed protocol is also optimal, in the sense that, the consistent global checkpoint taken in rVsAll protocol contains the minimal number of operations indispensable to fulfill the required session guarantees after server’s failure and its recovery. The paper presents the proof of the optimality property of the proposed protocol.

In the distributed computing environment, many large-scale scientific applications are irregular applications which perform their computation and I/O on an irregularly discretized mesh. However, most of the previous work in the area of irregular applications focuses mainly on the local environments. In distributed computing environments, since many remotely located scientists should share the data to produce useful results, providing a consistent data replication mechanism to minimize the remote data access time is a critical issue in achieving high-performance bandwidth. We have developed a replication software architecture(RSA) that enables the geographically distributed scientists to easily replicate irregular computations with minimum overheads, while safely sharing large-scale data sets to produce useful results. Since RSA uses database support to store the data-related and computational-related metadata, it can easily be ported to any computing environments. In this paper, we describe the design and implementation of RSA for irregular applications and present performance results on Linux clusters.

Job scheduling on computational grids is a key problem in large scale grid-based applications for solving complex problems. The aim is to obtain an efficient scheduler able to allocate dependable jobs originated from large scale applications on hierarchy based grid computing platforms with heterogeneous resources. In contrast to satisfying multi objectives of different levels, which is NP-hard in most formulations, a set of cooperative multi-objective genetic algorithm (MOGA) is presented. Using this scheme, the application level generates multiple local schedules based on local nodes and objectives to a schedule pool, from which the system level can assemble a set of global schedules according to global objectives. The MOGA scheduling scheme is shown to perform well on the experimental scenario, which shows its flexibility and possible application to more complex job scheduling scenarios with multiple and diverse tasks and nodes.

Several works on grid computing have been proposed during the last few years. However, most of them including available software, can not deal properly with some issues related to abstraction and friendliness of grid. A much wider variety of applications and large community of users can get benefit once grid computing technologies become easier to use and more sophisticated. This work concentrates on presenting ‘Users-Grid’ (pro-middleware), which sits between the grid middleware and user applications. It introduces a high-level abstraction layer and hides the intricacies of the grid middleware. It attempts to make the grid-related aspects transparent to the end users. Its main features are automatic DAG inference and seamless job submission. It facilitates end-users to directly run their applications on grid by themselves without depending on the expertise of support teams.