Papers by Jean-Marc Pierson
This paper presents the parallel implementation of a dynamic system of particles on a distributed... more This paper presents the parallel implementation of a dynamic system of particles on a distributed memory architecture. The implementation simulates the temporal evolution and behavior of a general system of particles. Long range interactions with the environment as well as short range interactions between particles are fairly handled. Distribution of the particles is a spaced-based partition, and the allocation of the particles on a linear network of processors is dynamic. An important study on the compromise between the cost of an unbalanced world and those of load balancing has been developed for a simple case, while an heuristic is proposed for general cases. Some results on MIMD architectures and on a network of workstations are nally presented.
Uniform Distributed Cache Service for Grid Computing
... The generated information is analyzed and used to manage the system dynamically. Therefore th... more ... The generated information is analyzed and used to manage the system dynamically. Therefore the uniform cache system applies the techniques of collaborative caching for data management in the grid. 1. Introduction ... 3. Collaborative Caching and Grid Caching ...
Medical data represent tremendous amount of data for which automatic analysis is increasingly nee... more Medical data represent tremendous amount of data for which automatic analysis is increasingly needed. Grids are very promising to face today's challenging health issues such as epidemiological studies through large image data sets. However, the sensitive nature of medical data makes it difficult to widely distribute medical applications over computational grids. In this paper, we review fundamental medical data manipulation requirements and we propose a distributed data management architecture that addresses the medical data security and high performance constraints. A prototype is currently being developed inside our laboratories to demonstrate the architecture capability to face realistic distributed medical data manipulation situations.
… of the 1st conference on Computing …, Jan 1, 2004
Pervasive computing applications allow users to access information from anywhere while traveling ... more Pervasive computing applications allow users to access information from anywhere while traveling and using variety of devices. Heterogeneity and limitation of resources involved in this application demand adaptation of content according to the current context (device, user, network etc.). The dynamic nature of adaptation mechanisms together with emerging opportunities of Web Service technology provides new approach of adaptation which is service-based. While this approach would provide a valuable service for the end customer, the service provider, and the content provider, it is important to have an architectural framework which is simple, scalable, flexible and interoperable. Moreover, in order to provide a complete service-based content negotiation and adaptation solution, we must have a model, or a tool, that allows defining environmental constraints, mapping them to appropriate adaptation service requirements and finding an optimal service configuration.
The question of energy savings has been a matter of concern since a long time in the mobile distr... more The question of energy savings has been a matter of concern since a long time in the mobile distributed systems and battery-constrained systems. However, for large-scale nonmobile distributed systems, which nowadays reach impressive sizes, the energy dimension (electrical consumption) just starts to be taken into account.
Future Generation …, Jan 1, 2012
In this paper we study the problem of energy-aware resource allocation for hosting long-term serv... more In this paper we study the problem of energy-aware resource allocation for hosting long-term services or on-demand compute jobs in clusters, e.g., deployed as part of computing infrastructures. We formalize the problem as three constrained optimization problems: maximize job performance under power consumption constraints, minimize power consumption under job performance constraints, and optimize a linear combination of power consumption and job performance. These problems are NP-hard but, given an instance, a bound on the optimal solution can be computed via a rational linear program. We propose polynomial heuristics for all three problems. Simulation experiments show that in all three cases some heuristics can achieve results close to optimal, i.e., lead to good job performance while conserving energy.
Proceedings of the 1st …, Jan 1, 2010
This paper presents an integrated framework for energy savings in large scale distributed systems... more This paper presents an integrated framework for energy savings in large scale distributed systems such as grids and clouds. The framework comprises tools and mechanisms: to measure and log data about the energy consumed by resources; to present this information to users; to involve the users in decisions to reduce the energy consumed by their applications; and to enforce energy reduction decisions automatically while respecting the users' requirements and achieving the resource availability demanded by current services. Experiments demonstrate the energy savings achieved by the proposed mechanisms and explore trade-offs between energy efficiency and performance degradation.
… of the 2009 international conference on …, Jan 1, 2009
There are a number of models for access control in pervasive environments that are based on trust... more There are a number of models for access control in pervasive environments that are based on trust propagation. Iterative multiplication of the trust values on a path from a source entity to a target entity is one of the common strategies for trust propagation. In this paper, we evaluate the effectiveness of iterative multiplication for trust propagation. The data set used for this evaluation is the real web of trust of Advogato.org that comprises of over 11, 000 vertices (users) and over 50, 000 directed weighted edges (trust relationships between users). We find that a significantly strong positive linear correlation exists between trust values based on direct experience and the corresponding propagated trust values derived through the iterative multiplication approach. This finding provides empirical support for the access control models for pervasive environments that employ the iterative multiplication strategy for trust propagation.
Euro-Par 2003 Parallel Processing, Jan 1, 2004
Access control is the field of security which deals with permissions to access resources, where r... more Access control is the field of security which deals with permissions to access resources, where resources may be computing power, storage capacity and data. On the other hand computational grids are systems, where users share those resources in a mostly transparent way. Grid access control poses novel challenges, since the distributed nature of grids make it difficult to manage access control by a central authority. Numerous overlapping domains with different access control policies exist and the sharing of storage resources makes it possible that data leaves the domain of its owner. To enable the owner to enforce his access control policy in such cases, access control solutions adapted to grid environments are needed. In this article we introduce Semantic Access Certificates as an extension to existing access control solutions for grids, to solve some problems that arise when grids are used to process medical data.
Proceedings of the first …, Jan 1, 2004
Computation and data grids have encountered a large success among the scientific computing commun... more Computation and data grids have encountered a large success among the scientific computing community in the past few years. The medical imaging community is increasingly aware of the potential benefit of these technologies in facing today medical image analysis challenges. In this paper, we report on a first experiment in deploying a medical application on a large scale grid testbed. Our pilot application is a hybrid metadata and image content-based query system that manipulates a large data set and for which image analysis computation can be easily parallelized on several grid nodes. We analyze the performances of this algorithm and the benefit brought by the grid. We further discuss possible improvements and future trends in porting medical applications to grid infrastructures.
Methods of information in medicine, Jan 1, 2005
In this article we present grids as an architecture for medical image processing and health-care ... more In this article we present grids as an architecture for medical image processing and health-care networks. We argue that confidential patient data should not be stored on the grid unprotected and explain why access control systems do not offer sufficient protection alone. Effective protection can be achieved by storing confidential data in encrypted form. Our proposal details a key management architecture, that allows encrypted storage and still enables users to access decryption keys for data they are authorized to see. Furthermore our architecture is robust against breakdowns, and denial of service attacks. It scales well with the number of users and does not introduce a single point of failure into the system.
Confidential data stored on mass storage devices is at risk to be disclosed to persons getting ph... more Confidential data stored on mass storage devices is at risk to be disclosed to persons getting physical or administrator access to the device. Encrypting the data reduces this risk, at the cost of more cumbersome administration. In this publication, we examine the problem of encrypted data storage in a grid computing environment, where storage capacity and data is shared across organizational boundaries. We propose an architecture that allows users to store and share encrypted data in this environment. Access to decryption keys is granted based on the grids data access permissions. The system is therefore usable as an additional security feature together with a classical access control mechanism. Data owners can choose different tradeoffs of security versus efficiency. Storage servers need not to be trusted and common access control models are supported.
Pervasive systems are created by introducing wireless communication into distributed multimedia s... more Pervasive systems are created by introducing wireless communication into distributed multimedia systems. These systems facilitate mobile data access to applications such as health care, tourism and emergency. However new challenges have emerged that has not been adequately addressed in traditional distributed multimedia systems. In pervasive systems, data must be adapted to the user's preferences, device capabilities and network conditions. Several content adaptation techniques have been proposed to fit the data to the usage environment. In this paper we present a distributed content adaptation framework (DCAF) for distributed multimedia systems and its prototype implementation. The architecture takes into consideration client profile, network conditions, content profile (meta-data) and available adaptation services (third party software elements) to construct an optimal dynamic adaptation graph.
From Grid to Healthgrid, …
This paper presents µgrid, a light weight middleware for grid applications, and focuses mainly on... more This paper presents µgrid, a light weight middleware for grid applications, and focuses mainly on security issues -more specifically on the access control to resources -that are critical for the gridification of many medical applications. For this purpose, we use Sygn as a distributed, certificate based, and flexible access control mechanism, which has been fully integrated in µgrid. We discuss the advantages of the solution compared to classical grid approaches and the limitations of the final architecture.
… and Computation: Practice …, Jan 1, 2007
Distributed systems like grids support diverse models of distributed computation and need to oper... more Distributed systems like grids support diverse models of distributed computation and need to operate large data entities in a distributed way. A significant quantity of this data are used only for a limited period of time. Caching is recognized as one of the most effective techniques to manage temporary data and collaborative cache is traditionally proposed to scale cache capabilities in distributed environments. Grid needs to manage dynamically different models of computation with different data access patterns. In this paper, we propose a basic infrastructure for the management of collaborative caches that permits to operate and control dynamically different cache mechanisms and cache schemes in grid. Beside traditional collaborative caching where the cooperation is often limited to data resolution, in our infrastructure the collaborative cache capacities are extended to operate and manage these distributed temporal data. Our proposition is composed of a reference cache model that defines four layers for the management of collaborative cache; an information model that represents the main cache elements and their activity; and a set of operations to request specific tasks to monitor, operate, and coordinate a generic collaborative cache system. Implementation issues of a prototype in Globus Toolkit 4 are discussed.
A thin client architecture for data visualization on the world wide web
… Conference on Visual Computing (ICVC99), Goa, India, Jan 1, 1999
Third International IEEE …, Jan 1, 2007
Trust based access control models have recently attracted significant interest in the area of per... more Trust based access control models have recently attracted significant interest in the area of pervasive computing. In several trust based models, organizations are required to establish a set of trust beliefs regarding their neighboring nodes. In an organization there may be multiple administrators or the administrators may change with time. When multiple administrators contribute to the creation of the set of trust beliefs of an organization, inconsistencies may occur due to variations in their disposition to trust.
trust Certification Model for Large Access in Pervasive Environment
JPCC International Journal of Pervasive Computing …, Jan 1, 2005
Information grids: managing and mining semantic data in a grid infrastructure; open issues and application to geno-medical data
Page 1. Proceedings of the 14th International Workshop on Database and Expert Systems Application... more Page 1. Proceedings of the 14th International Workshop on Database and Expert Systems Applications (DEXA'03) 1529-4188/03 $17.00 © 2003 IEEE Page 2. Proceedings of the 14th International Workshop on Database and ...
In this paper we present the metadata usage in a medical imaging project grid. Metadata represent... more In this paper we present the metadata usage in a medical imaging project grid. Metadata represent data about the data: In our case, the data are medical images and the metadata store relative information on the patient and hospital records, or even data about the image algorithms used in our application platform. Metadata are either static or dynamically constructed after computations on data. We show how the metadata is used, produced and stored to provide a secure and efficient access to medical data (and metadata) through a dedicated architecture. Experiments include times to access data and to secure the transactions.
Uploads
Papers by Jean-Marc Pierson