Majed Abdouli

Scheduling distributed real-time nested transactions

To ensure global serializability, traditional distributed database systems implement both concurrency control and transaction commit protocols. In a distributed database system, a commit protocol guarantees the uniform commitment of distributed transaction execution. In the last decade, several extensions to the transaction model adopted in traditional database systems have been proposed in order to support the functional and performance requirements of emerging advanced applications such as CAD/CAM, large software design projects and object-oriented databases. Nested transaction models have been shown to play an important role in such applications, however, these models are not yet fully studied. In this paper, our contributions in the field of real-time nested transactions are two fold: (i) we propose a hierarchical and flat protocols for real-time nested transactions, called 2PC-RT-NT, and (ii) we implement a lock mechanism, called 2LP-NT-HP, to solve the data conflicts problem between nested transactions.

Intelligent framework for traceability of containerized goods

In recent years, the work of research in the field of traceability is led to the development systems used to draw effectively the information on the level the supply chain. With a focus on the importance of the concept of the traceability, this article describes our proposal which is summarized in a smart based architecture the linkage of paradigm agent and semantic web while ensuring the traceability of containerized goods in a context of multimodal transport. The content is organized in the following way: the first part is devoted to a state of the art as regards the management of traceability in the supply chain. Then, we submitted our contribution and by which we have introduced the concepts MAS (Multi Agent System) and semantic web to assimilate properly our system i-TSCG (intelligent Traceability System Containerized Goods) composed of five layers. Thus, we chose the methodology MASE (Multi-Agent System Engineering) for the design and implementation of our system.

A system supporting nested transactions in DRTDBSs

Extended transaction models in databases were motivated by the need of complex applications such as CAD/CAM and software engineering. Nested transaction models have so far been shown to play an important role in such applications. However, these models are not yet fully studied. In this paper, we focus on the applicability of such models to real-time database systems, particularly to issues related to the global serializability of distributed real-time nested transactions. Our contribution in this field is twofold: we propose (i) a real-time concurrency control, called 2PL-NT-HP, to solve data conflicts problem between nested transactions and (ii) a real-time commit protocol to guarantee the uniform commitment of distributed nested transactions. To this purpose, we have adapted the PROMPT real-time commit protocol which is designed specifically for the real-time flat transactions. This protocol causes intra-aborts cascade in nested environment and hence decreases its real-time performances. To alleviates this drawback, the borrowing subtransaction carries out a speculative execution by accessing both before and after-image of the lending subtransaction. Simulations we have carried out show that S-PROMPT approach is very useful in DRTDBSs compared to the classical approaches.

Multi-hop Energy Efficient routing protocol based on Data Controlling for Wireless Sensor Networks

Wireless Sensor Networks (WSN) differ from traditional wireless communication networks in several characteristics. One of them is the power awareness since the batteries of sensor nodes have a restricted lifetime and are difficult to be replaced. Therefore, all protocols must be designed to minimize energy consumption and to preserve the longevity of the network. Routing protocols in WSN aim at accomplishing power conservation. Most of researches have focused on energy efficient solutions regardless of the communication between Cluster Heads (CH) and Base Station (BS). When the sensor field is far away from the BS, the CH are burdened with heavier relay traffic and tend to die much faster. To solve this problem, a Multi-hop Energy Efficient routing protocol based on Data Controlling (MEEDC) is proposed in this paper. Our work focuses on (i) a heterogeneous network and (ii) a sensitive data controlling. The role of super nodes in the proposed heterogeneous network is to connect the CH and the BS. They are responsible for transmitting packets received from the CH to the BS, thus the CH can preserve some energy in data forwarding. The control of data transmission tries to reduce the number of transmissions and thus considerable energy conservation is achieved. Simulation results show that the MEEDC protocol significantly prolongs the network survival time.

Event-driven wireless sensor networks based on consensus

Wireless Sensor Networks (WSN) consist of sensor nodes with sensing and communication capabilities. Since sensor nodes are energy constrained and sensor data are subject to several sources of errors, the use of a quorum mechanism to achieve consensus leads to higher data accuracy. In this paper, an Event Clustering Routing Protocol based on Consensus (ECRPC) is proposed for WSN. The main idea of ECRPC is based on: (i) No message will be transmitted to the Base Station (BS) provided that a consensus is reached (ii) A tradeoff is made between the two criteria of residual energy and link cost: in certain cases, we give priority to length relay over the energy or vice versa. Simulation results show that ECRPC protocol significantly prolongs the network survival time.

A new hybrid routing protocol for Wireless Sensor Networks

Wireless sensor networks (WSN) differ from traditional wireless communication networks in several characteristics. One of these characteristics is power awareness. In order to save the overall energy of the system and to fairly balance the load among nodes, we propose a new clustering routing protocol based on coverage control which incorporates two modules: (1) a scheduling mechanism based on binary particle swarm optimisation (PSO); (2) a cluster based protocol using fuzzy C-means (FCM) algorithm. Yet, in cluster based approach, the cluster head (CH) is usually selected amongst the sensor nodes which can die quickly for this extra workload. To mitigate this problem, we introduce some energy-harvesting (EH) nodes in the network which serve as CH. Since, the EH nodes are more expensive than the sensors, we study, in this paper, how to efficiently deploy EH nodes. Simulation results show that our protocol achieves an obvious improvement on the network lifetime.

A new fault-tolerant architecture based on DASH for adaptive streaming video

The standard dynamic adaptive streaming over HTTP (DASH) has been proposed to guarantee a quality of video under varying bandwidth. DASH enables adaptation of the media bit-rate to varying throughput conditions by offering multiple representations of the same content. However, the Moving Picture Experts Group (MPEG) DASH specification only defines the media presentation description (MPD) and the segment formats. In order to take advantages of DASH, we propose a fairness architecture, called dynamic adaptive streaming over HTTP distributed multimedia systems (DASH-DMS). DASH-DMS combines both 2-tiers and 3-tiers architectures in order to guarantee a quality of service (QoS) under changing conditions. Moreover, we propose replication algorithms to enable load balancing video servers and improve the global QoS. Furthermore, in order to improve the availability of our system, we use a fault tolerance policy. To this purpose, we have conduct simulations throughout the paper using the simulator developed by our team.

Video Replication over a New Architecture DASH-DMS

Nowadays, the video traffic represents more than 65% of global traffic and in the Internet forecasts, it will reach 79% in 2018 in both wired and wireless environments. Hence, major content providers such as Netflix, Youtube, Hulu, and Vudu are leveraging HTTP-based multimedia transmission with adaptative streaming with different solution in order to guarantee a quality of service (QoS). Moreover, in 2012, a new standard called Dynamic Adaptive Streaming over HTTP (DASH) which enables adaptation of the media bitrate to varying throughput conditions by offering multiple representations of the same content is proposed. In this paper, we present a fairness architecture, called DASH-DMS, in combination with the new Dynamic Adaptive Streaming over HTTP (DASH) standard, in order to guarantee a certain QoS under changing conditions in the available bandwidth. DASH-DMS is a hybrid architecture which combines both 2-tiers and 3-tiers architectures. Moreover, we present a replication algorithm to enable load balancing video servers and improve the global QoS. Furthermore, in order to improve the availability of our system, we use a fault tolerance policy. Simulations conducted along this paper show that our proposition significantly outperforms existing and state-of-the-art approaches.

DASH-DMS: To improve streaming video over HTTP

Recently, as shown by different studies, the video traffic dominates the current traffic composition of the Internet. Nowadays, the video traffic represents more than 65% of global traffic and in the Internet forecasts, it will reach 79% in 2018 in both wired and wireless environments. Hence, major content providers such as Netflix, Youtube, Hulu, and Vudu are leveraging HTTP-based multimedia transmission with adaptative streaming with different solution in order to guarantee a quality of service (QoS). Moreover, in 2012, a new standard called Dynamic Adaptive Streaming over HTTP (DASH) which enables adaptation of the media bitrate to varying throughput conditions by offering multiple representations of the same content is proposed. In this paper, we present a fairness architecture, called DASH-DMS, in combination with the new Dynamic Adaptive Streaming over HTTP (DASH) standard, in order to guarantee a certain QoS under changing conditions in the available bandwidth. DASH-DMS is a hybrid architecture which combines both 2-tiers and 3-tiers architectures. Moreover, we present a replication algorithm to enable load balancing video servers and improve the global QoS. Simulations conducted along this paper show that our proposition significantly outperforms existing and state-of-the-art approaches.

F-RT-ETM: toward analysis and formalizing real time transaction and data in real-time database

Due to the diversity of extended transaction models, their relative complexity and their lack of formalization, the characterization and the comparison of these models become delicate. Moreover, these models capture only one subset of interaction which can be found in the spectrum of the possible interactions. In front of this established fact, the framework ACTA was introduced. Our contribution in this field is twofold: (i) we extend ACTA by adding many dependencies for capturing a new interaction between transactions in real time environment, and we extend ACTA to take into account temporal characteristics of real-time data item (ii) we presented a meta-model that capture concept of an extended real time transaction model by using UML class diagram and its formal description using Z language.