Catálogo de publicaciones - libros

Distributed applications are usually concurrent and nondeterministic. For this reason, formal verification on their design specifications is an essential technique for us to gain more confidence in the correctness of the behavioral aspects of our design before putting them into coding stage. Message-Oriented Middleware (MOM) is widely used to simplify the task of interprocess communications in distributed applications. To model the MOM-based applications for verification purpose, the services provided by MOM must also be integrated into the models. However, MOM modeling is non-trivial. While providing high-level program interfaces which shield programmers from the complexity of the underlying operating systems and networks, MOM may also conceals under such interfaces the concurrency and nondeterminism present in the underlying networks. This increases the possibility of misinterpretting the behavior of the applications, which in turn causes design errors. An over-abstracted MOM model based ons Application Programming Interface may bury such design errors while an over-detailed model may consume too much resource and render the verification infeasible. As a guideline for MOM modeling, we present several formal models of various behavioral aspects of MOM in terms of Promela, the specification language used in SPIN model checker. Based on our empirical study, we also discuss the impact of incorporating these formal models in different settings into the MOM-based application models, in terms of increased state space for model checking.

One of the major challenges for grid technologies is to create the scientific and technological base for share, collaboration, large-scale distributed systems. Theories and models of grid architectures are important to this endeavor as well as to providing the foundations for constructing grid systems able to work effectively. It is important to model and analyze the grid architecture so that it can evolve guided by scientific principles. On the basis of a coarse-grain classification of grid applications, we present a novel grid architecture taxonomy, interaction-intensive and computation-intensive architecture. In this paper, we will give some new grid performance metrics and model grid architectures mathematically via queueing networks. In addition, we obtain some scientific principles guiding the grid architecture design.

A peer community is composed of a number of peers who share files about the same topic in the file sharing P2P applications. Building peer communities can benefit content location and retrieval in P2P systems. We propose an effective approach based on rough set and topic cluster to build peer communities. Firstly, we compute one of the best reduced sets of all the same type files, such as the video files, with files’ attributes in a peer. Secondly, topic clusters of a peer are calculated, which represent the interests of it. Finally, we build peer communities using the super peer technique. Experiments performed on the real data sets prove that our approach is effective. Experimental results verify that our approach works much better compared with that of previous approaches.

This paper describes the performance evaluation to validate the superiority of UbiMDR framework. In ubiquitous application, the semantic operability is one of the most important issues to maximize the usability of sensors in sensor fields. However, existing frameworks are not suitable for the ubiquitous computing environment because of data heterogeneity between data elements. The MDR-based framework in ubiquitous computing provides the semantic interoperability among ubiquitous applications or sensor fields. In addition, the UbiMDR framework represents low costs for mapping or addition of new elements or operation compared to conventional framework.

Heterogeneous Networks of Workstations (NOWs) offer a cost-effective solution for parallel processing. The completion time of a parallel task over NOWs depends on how the task is divided and distributed among the heterogeneous workstations. In this paper we present a distribution scheme which attempts to minimize the task’s completion time over a heterogeneous NOWs. The scheme is based on the idea of distributing fixed time slices of work as opposed to fixed work slices. Our simulations show that the proposed scheme outperforms both fixed and variable work distribution schemes commonly in use. The scheme is very simple and requires no active monitoring of the network. Furthermore it is adaptive and copes very well with the changes in background loads on workstations and network interference.

In this study, we model pricing of grid/distributed computing resources as a problem of real option pricing. Grid resources are non-storable compute commodities (eg., CPU cycles, memory, etc). The non-storable characteristic feature of the grid resources hinders it from fitting into a risk-adjusted spot price model for pricing financial options. Grid resources users pay upfront to acquire and use grid compute cycles in the future, for example, six months. The user expects a high and acceptable degree of satisfaction expressed as the Quality of Service (QoS) assurance. This requirement further imposes service constraints on the grid because it must provide a user-acceptable QoS guarantee to compensate for the upfront value. This study integrates three threads of our research; pricing the grid compute cycles as a problem of real option pricing, modeling grid resources spot price using a discrete time approach, and addressing uncertainty constraints in the provision of QoS using fuzzy logic. We have proved the feasibility of this model through experiments and we have presented some of our pricing results and discussed them.

To gain high performance computing or store large amount of data using inexpensive devices, grid system is one of the well-known solutions. In most cases, the grid can be categorized into two types: computational grid and data grid. Data grid is used for data intensive applications. In data grids, replication is used to reduce access latency and bandwidth consumption. Furthermore, it can also improve data availability, load balancing and fault tolerance. If there are many replicas, they may have coherence problems while being updated. In this paper, based on the aggressive-copy method, we propose a novel Greedy Pipeline-based Aggressive Copy (GPAC) protocol. The performance of pipelining dataset blocks and greedy sequencing in the GPAC can accelerate data replication speed in compared with previous works. Both analytical and experimental results show promising performance enhancements.

The performance of contiguous allocation strategies can be significantly affected by the distribution of job execution times. In this paper, the performance of the existing contiguous allocation strategies for 3D mesh multicomputers is re-visited in the context of heavy-tailed distributions (e.g., a Bounded Pareto distribution). The strategies are evaluated and compared using simulation experiments for both First-Come-First-Served (FCFS) and Shortest-Service-Demand (SSD) scheduling strategies under a variety of system loads and system sizes. The results show that the performance of the allocation strategies degrades considerably when job execution times follow a heavy-tailed distribution. Moreover, SSD copes much better than FCFS scheduling strategy in the presence of heavy-tailed job execution times. The results also show that the strategies that depend on a list of allocated sub-meshes for both allocation and deallocation have lower allocation overhead and deliver good system performance in terms of average turnaround time and mean system utilization.

Some challenges in the mobile grid environments are not handled by some related works, for example, adapting to heterogeneous interfaces of different mobile devices (PDAs and cellular) for submission and monitoring of applications in grid environments. This article presents an approach that employs the workflow concept for providing automated and adapted features for executing applications in grid mobile configurations. The approach, coined as SuMMIT, enabled to consume less battery energy of PDAs and more agility for submitting and monitoring applications in comparison with some related works. In addition, the SuMMIT environment provides an execution flow adjustment, in case of disconnection, matching requirements of submitted application and options defined by the user. The SuMMIT also has adapted and optimized characteristics according to some limitations and problems found in different devices.

Grids are becoming the platform of choice for high performance computing. Although grids present a unified view of resources, they need evolved user interfaces in order to fully take advantage of their potential for real applications. Grid portals can deliver complex grid solutions to users; they do so without the need to download or install specialized software, or worrying about setting up networks, firewalls, and port policies. Due to the powerful, general-purpose nature of grid technology, and the nature of the grid resources they expose, the security of portals or points of access to such resources must be carefully considered. In this paper we present , the Grid Portal, a user-specific portal which allows Java applications to access grid resources for execution. We describe how the portal exploits standard, off-the-shelf commodity software together with existing grid infrastructures in order to facilitate security and data access. The main technologies used by are GSI, MyProxy, Java CoG Kit, GridSphere, and middleware. In , Java classes and files are loaded on demand from the user’s machines. We describe how the Grid Portal supports this execution model and how it allows users to access controlled external data servers (users’ file systems and file systems accessible from their local workstations), under a secure platform.