Catalin Leordeanu

A hybrid algorithm for scheduling workflow applications in grid environments (ICPDP)

In this paper, based on a thorough analysis of different policies for DAG scheduling, an improved algorithm ICPDP (Improved Critical Path using Descendant Prediction) is introduced. The algorithm performs well with respect to the total scheduling time, the schedule length and load balancing. In addition, it provides efficient resource utilization, by minimizing the idle time on the processing elements. The algorithm has a quadratic polynomial time complexity. Experimental results are provided to support the performance evaluation of the algorithm and compare them with those obtained for other scheduling strategies. The ICPDP algorithm, as well as other analyzed algorithms, have been integrated in the DIOGENES project, and have been tested by using Mon-Alisa farms and ApMon, a MonAlisa extension.

A solution for the management of multimedia sessions in hybrid clouds

Cloud-based systems expanded considerably in recent years, as the demand for cheaper and easily scalable resources provisioning solutions increased. In this context, the deployment of multimedia services in the cloud, as a way to increase their usability and overcome their processing overhead, gained additional interest. Throughout this paper we present the design of a management system for multimedia services built on top of a hybrid cloud. This describes a model of service unit deployment automation as well as a services composition scheme adapted to the IaaS cloud architecture and resources pool. We propose a modular and flexible architecture to handle service composition, automated service deployment with execution flow management and scalable storage capabilities. Finally we analyse the benefits of using the service management system with a hybrid cloud back-end.

A fault tolerance approach for distributed systems using monitoring based replication

High availability is a desired feature of a dependable distributed system. Replication is a well-known technique to achieve fault tolerance in distributed systems, thereby enhancing availability. We propose an approach relying on replication techniques and based on monitoring information to be applied in distributed systems for fault tolerance. Our approach uses both active and passive strategies to implement an optimistic replication protocol. Using a proxy to handle service calls and relying on service replication strategies, we effectively deal with the complexity and overhead issues. This paper presents an architecture for implementing the proxy based on monitoring data and the replication management. Experimentation and application testing using an implementation of the architecture is presented. The architecture is demonstrated to be a viable technique for increasing dependability in distributed systems.

A dependability layer for large-scale distributed systems

Ensuring dependability in large-scale distributed systems represents an important research subject today. Despite the fact that many projects obtained valuable results in this domain, no acceptable solution was yet found that could integrate all the requirements for designing a dependable system and that could exploit all the capabilities of modern systems. We present a unitary and aggregate approach to ensuring reliability, availability, safety and security of distributed systems. Starting from the proposed architecture, we present implementation details for two solutions designed to ensure fault tolerance, using virtualisation and container-based replication of services. We also present an approach to enhance security using combined modern security models in large-scale distributed systems. The results and implementation details can serve as a methodology to assist distributed infrastructures in adopting such a middleware layer designed to enforce dependability in large-scale distributed systems.

Policy Based Resource Allocation in Cloud Systems

Cloud Systems provide computing resources in a flexible manner. There are several key requirements that need to be addressed regarding the resource allocation in Clouds and the most important of them is providing on demand elasticity. This paper focuses on adding new features to the Cloud resource allocation mechanism that enhance on demand elasticity. Most of the resource managers that are now on the market use static allocation. We propose a novel solution that uses dynamic allocation, based on well defined policies. Moreover, the proposed solution offers authentication and accountability for the actions of the users which is very important for the commercial aspect of public clouds.

Secure Access to Cloud Resources

Cloud computing has developed into an ubiquitous distributed system model, offering services which range from storage to computing power or database infrastructure. The security aspect of such distributed systems is a very important and challenging issue. In this paper we propose a security solution which offers policy enforcement in order to detect denial of service attacks and to manage quota parameters of users registered in a cloud. We also present novel methods for controlling access suitable for cloud computing platforms which have high security requirements. Because cloud computing is a dynamic environment where the resources are allocated on demand and the users come and leave, a different approach must be taken. We analyze different access control models like Mandatory Access Control, Discretionary Access Control and Role-Based access control models and propose a model adapted to the Cloud Computing environment. Moreover, we investigate the cloud access control security issues and build a role-based access model that meets the cloud requirements. The proposed model inherits Role-Based Access Control features, but also has other features like temporal rights and constraints.

Netpy: Advanced Network Traffic Monitoring

The paper presents an extension of NetPy, a network traffic visualization tool using NetFlow records. NetPy offers traffic visualization as well as a traffic analysis, which is very useful for the network administrators. NetPy has been re-designed to include a better Graphical User Interface to make the application easier to use and to present the information in the best way possible. The new version also improves the former analysis modules and offers new ways for the administrator to monitor and control events in the network, such as a comparative analysis. The application also integrates a new advanced intrusion detection module. The attacks that this module is able to detect will be presented, as well as the methods used. The Denial of Service, PortScan and custom defined attacks, based on the number of connections, the number of packets or the payload between hosts can be detected with a very good accuracy. The new NetPy version has been developed by a team from Politehnica University of Bucharest in collaboration with the department of Computer Science of the University of Wisconsin-Madison U.S.A.

Correlation of Intrusion Detection Information in Grid Environments

This paper presents a novel intrusion detection system for Grid Systems. It is intended to identify potential attackers who try to modify or compromise the applications sent to execution by various users or target different resource groups within the Grid. The system makes use of a number of available local intrusion detection systems which send data to a grid-level intrusion detection system that takes decisions based on an overview of the entire Grid. This IDS can correlate the information received from the local systems, as well as monitoring data from the Grid System, using statistical methods, to identify attacks that cannot be detected at a local level. Another contribution of this paper is the classification of threats based on the intent of the attacker. This paper also demonstrates that these types of attacks can be detected using the proposed complex intrusion detection system.

Integrated Cloud Framework for Farm Management

Smart or intelligent farming is a relatively new concept, but its becoming an important factor for the agricultural sectors, as a way to raise productivity through technology. This article presents a global management and decision system for smart farming. Our system is addressed to farmers and other actors involved in farming operations, such as rural service companies and banks. It can manage up to several groups of smart or technology-based farms. The system is functioning on two levels, local and Cloud. First of all, it must be deployed on the farms local wireless network. And secondly, it uses the Clouds resources and flexibility to offer even more complex services.

Towards a Hybrid Local-Cloud Framework for Smart Farms

Smart or intelligent farming is a relatively new concept, but its becoming an important factor for the agricultural sectors, as a way to raise productivity through technology. This article presents a global management and decision system for smart farming. Our system is addressed to farmers and other actors involved in farming operations, such as rural service companies and banks. It can manage up to several groups of smart or technology-based farms. The system is functioning on two levels, local and Cloud. First of all, it must be deployed on the farms local wireless network. And secondly, it uses the Clouds resources and flexibility to offer even more complex services.