Catálogo de publicaciones - libros

This article presents , a parallel branch and bound framework for grids. Branch and bound (B&B) algorithms find optimal solutions of search problems and NP-hard optimization problems.

is a Java framework that helps programmers to distribute problems over grids by hiding distribution issues. It is built over a master-worker approach and provides a transparent communication system among tasks. This work also introduces a new mechanism to localize computational nodes on the deployed grid. With this mechanism, we can determine if two nodes are on the same cluster. This mechanism is used in to reduce inter-cluster communications. We run experiments on a nationwide grid. With this test bed, we analyze the behavior of a communicant application deployed on a large-scale grid that solves the flow-shop problem.

The CMS experiment will produce several Pbytes of data every year, to be distributed over many computing centers geographically distributed in different countries. Analysis of this data will be also performed in a distributed way, using grid infrastructure. CRAB (CMS Remote Analysis Builder) is a specific tool, designed and developed by the CMS collaboration, that allows a transparent access to distributed data to end physicist. Very limited knowledge of underlying technicalities are required to the user. CRAB interacts with the local user environment, the CMS Data Management services and with the Grid middleware. It is able to use WLCG, gLite and OSG middleware. CRAB has been in production and in routine use by end-users since Spring 2004. It has been extensively used in studies to prepare the Physics Technical Design Report (PTDR) and in the analysis of reconstructed event samples generated during the Computing Software and Analysis Challenge (CSA06). This involved generating thousands of jobs per day at peak rates. In this paper we discuss the current implementation of CRAB, the experience with using it in production and the plans to improve it in the immediate future.

Resource Allocation in Grid environments to date is generally carried out under the assumption that there is one primary scheduling system scheduling jobs. However, as environments tend towards larger open “utility” Grids it becomes increasingly likely that deployments will involve multiple independent schedulers allocating jobs over the same resources.

In this paper we show that, if using current standard scheduling approaches, such multi-scheduler environments may well be prone to serious oscillation problems in resource allocation similar to those commonly found in IP network traffic. Further we demonstrate how common techniques from IP networks – in particularly approaches based on Random Early Detection (RED) buffer management and its subsequent extensions / variations – may provide an effective way to damp or eliminate such oscillations. The paper describes the analogy between multi-scheduler Grid resource allocation and IP network routing and explores the impact of oscillation and RED methods by simulation.

The vast computing potential of mobile computing systems is often hampered by their susceptibility to transient and independent failures. To add reliability and high availability to such systems, checkpoint based rollback recovery is one of the widely used ones for scientific computing, database, telecommunication and mission critical applications. This paper presents a coordinated nonblocking checkpointing and recovery technique for such systems that handles the constraints posed by the underlying wireless network, efficiently. Here an initiator (an MSS) sends checkpoint requests to all other MSSs and the MSSs send this request only to those MHs, which have communicated in the last checkpointing interval (relieving the wireless network from synchronization overhead). Also all acknowledged messages are logged at the home station of the receiver MH so that only the faulty MHs need to recover in case of failure and no other process is affected by this fault and subsequent recovery.

Service discovery is an integral part of Mobile Ad Hoc Networks (MANETs). While several service discovery protocols such as Service Location Protocol [1] and Universal Plug and Play (UPnP)[2] have been developed, most of them are designed for infrastructure based networks and thus not suitable to be used in MANETs. This paper proposes Mobile Agent Based Service Discovery (MASD) for Ad Hoc network. It is a policy driven agent based mechanism that facilitates cross platform service discovery in ad-hoc environments. The various agents are chosen in MASD which executes the predefined policies. Our approach achieves a high degree of flexibility in adapting itself to changes in ad-hoc environments and is aware of common problems associated with structured compound formation in mobile Ad Hoc environments. Mechanism consists of grouping mobile nodes into clusters while a gateway in each cluster is responsible for routing. We evaluated the performance of the scheme by running simulations on Glomosim simulator. Also the motion of agents is taken care of by PMADE (Platform for Mobile Agent Distribution and Execution).The results obtained shows that MASD is quite effective for successful service discovery in MANETs.

Wireless technology is becoming an attractive mode of communication for real-time applications in typical settings such as in an industrial setup because of the tremendous advantages it is capable of offering. However, the high bit error rate characteristics of wireless channel due to conditions like attenuation, noise, fading and interference seriously impact the timeliness and guarantees that need to be provided for real-time traffic. Existing wireless protocols either do not adapt well to the erroneous channel conditions or do not provide real-time guarantees. The goal of our work is to design and evaluate novel real-time MAC (Medium Access Control) protocols for scheduling messages in a 2-level hierarchical wireless industrial network taking into account the time-varying channel condition. Our objective is to minimize the loss rate of messages using the slot exchange protocol[9] that actively combats the erroneous channel conditions and maximize the channel utilization by enabling parallel transmissions in a collision-free manner. Unfortunately, these two goals have inherent conflicts in shared medium wireless networks. We propose a distributed protocol, called the Adaptive protocol that arbitrates between these two design criteria in order to resolve the inherent conflict between them. Through simulation studies, we show that the proposed Adaptive protocol achieves significant improvement in deadline miss ratio compared to the baseline protocols that exploit complete parallelism and full exchange, for a wide range of channel conditions.

Mobile agents are very useful in low bandwidth ad hoc network environments. But various attacks are possible against both agent and agent platforms. In this paper we have modeled hierarchical scheme for protecting mobile agent environment which is applicable to both wired and wireless networks. Colored Petri Net is used as a modeling tool which effectively models platform behavior and agent mobility through the movement of tokens and timed firing of transitions. We have considered centralized security scheme for our model where Trusted-Third-Party plays a central role in the communication.

In a mobile computing environment, users can perform on-line transaction processing independent of their physical location. In a mobile environment, multiple mobile hosts may update the data simultaneously which may result in inconsistency of data. To solve such problems many concurrency control techniques have been proposed. The traditional two phase locking protocol has some inherent problems such as deadlocks & long unpredictable blocking. In this paper we propose a concurrency control mechanism with dynamic timer adjustment which helps in reducing the communication overhead and enhances the transaction throughput. The simulation results specify the performance trade off metrics.